U.S. patent number 8,953,522 [Application Number 13/074,564] was granted by the patent office on 2015-02-10 for method and apparatus for controlling retransmission on uplink in a wireless communication system supporting mimo.
This patent grant is currently assigned to Samsung Electronics Co., Ltd.. The grantee listed for this patent is Jin-Kyu Han, Youn-Sun Kim, Myung-Hoon Yeon, Han-Il Yu. Invention is credited to Jin-Kyu Han, Youn-Sun Kim, Myung-Hoon Yeon, Han-Il Yu.
United States Patent |
8,953,522 |
Han , et al. |
February 10, 2015 |
Method and apparatus for controlling retransmission on uplink in a
wireless communication system supporting MIMO
Abstract
A method is provided for controlling retransmission by a User
Equipment (UE) in a wireless communication system supporting
Multiple Input Multiple Output (MIMO) technology. A plurality of
transport blocks is initially transmitted to a Node B. A
retransmission request for at least one transport block among the
plurality of transport blocks is received from the Node B. A
precoding matrix for retransmission of the at least one transport
block is determined based on the retransmission request for the at
least one transport block. The at least one transport block is
retransmitted using the determined precoding matrix.
Inventors: |
Han; Jin-Kyu (Seoul,
KR), Kim; Youn-Sun (Seongnam-si, KR), Yeon;
Myung-Hoon (Yongin-si, KR), Yu; Han-Il
(Seongnam-si, KR) |
Applicant: |
Name |
City |
State |
Country |
Type |
Han; Jin-Kyu
Kim; Youn-Sun
Yeon; Myung-Hoon
Yu; Han-Il |
Seoul
Seongnam-si
Yongin-si
Seongnam-si |
N/A
N/A
N/A
N/A |
KR
KR
KR
KR |
|
|
Assignee: |
Samsung Electronics Co., Ltd.
(KR)
|
Family
ID: |
44656416 |
Appl.
No.: |
13/074,564 |
Filed: |
March 29, 2011 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20110235586 A1 |
Sep 29, 2011 |
|
Foreign Application Priority Data
|
|
|
|
|
Mar 29, 2010 [KR] |
|
|
10-2010-0028207 |
Apr 19, 2010 [KR] |
|
|
10-2010-0036134 |
|
Current U.S.
Class: |
370/328;
370/310.2; 370/338 |
Current CPC
Class: |
H04L
1/0031 (20130101); H04L 25/03929 (20130101); H04L
1/1671 (20130101); H04B 7/0473 (20130101); H04B
7/0426 (20130101); H04L 25/0391 (20130101); H04B
7/0617 (20130101); H04L 1/1607 (20130101); H04L
25/03942 (20130101); H04L 1/1812 (20130101); H04L
25/03923 (20130101) |
Current International
Class: |
H04W
4/00 (20090101) |
Field of
Search: |
;370/328,338 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
101682382 |
|
Mar 2010 |
|
CN |
|
1 959 585 |
|
Aug 2008 |
|
EP |
|
2 141 852 |
|
Jan 2010 |
|
EP |
|
2007-116637 |
|
May 2007 |
|
JP |
|
2007-214824 |
|
Aug 2007 |
|
JP |
|
2009-164976 |
|
Jul 2009 |
|
JP |
|
2009-219116 |
|
Sep 2009 |
|
JP |
|
2010-500790 |
|
Jan 2010 |
|
JP |
|
2 330 381 |
|
Jul 2008 |
|
RU |
|
WO 2006/118433 |
|
Nov 2006 |
|
WO |
|
WO 2007/052941 |
|
May 2007 |
|
WO |
|
WO 2008/156067 |
|
Dec 2008 |
|
WO |
|
WO 2009/088167 |
|
Jul 2009 |
|
WO |
|
WO 2009/110759 |
|
Sep 2009 |
|
WO |
|
WO 2009/123522 |
|
Oct 2009 |
|
WO |
|
Other References
LTE; Evolved Universal Terrestrial Radio Access (E-UTRA); Physical
Channels and Modulation (3GPP TS 36.211 Version 9.0.0 Release 9).
cited by applicant .
LTE; Evolved Universal Terrestrial Radio Access (E-UTRA);
Physical(Jan. 2010) Channels and Modulation (3GPP TS 36.211 Version
9.0.0 Release 9). cited by applicant .
XP014045995, Jan. 1, 2010. cited by applicant.
|
Primary Examiner: Ton; Dang
Assistant Examiner: Kaur; Pamit
Attorney, Agent or Firm: The Farrell Law Firm, P.C.
Claims
What is claimed is:
1. A method for controlling retransmission by a User Equipment (UE)
in a wireless communication system supporting Multiple Input
Multiple Output (MIMO) technology, comprising the steps of:
initially transmitting a plurality of transport blocks to a Node B,
and receiving a negative acknowledgement (NACK) for at least one
transport block among the plurality of transport blocks, from the
Node B; determining a precoding matrix for retransmission of the at
least one transport block in response to the NACK for the at least
one transport block; and retransmitting the at least one transport
block using the determined precoding matrix, wherein the precoding
matrix is determined based on a number of layers corresponding to
the at least one transport block negatively acknowledged (NACKed)
from the Node B, if the UE does not receive a Physical Downlink
Control CHannel (PDCCH) intended for the UE from the Node B.
2. The method of claim 1, wherein the NACK is transmitted over a
Physical HARQ Indicator CHannel (PHICH).
3. The method of claim 2, wherein the PHICH comprises single
ACK/NACK information for a plurality of transport blocks, and if
the single ACK/NACK information is indicated as NACK information, a
default precoding matrix is used as the precoding matrix for
retransmission.
4. The method of claim 2, wherein the PHICH comprises single
ACK/NACK information for a plurality of transport blocks, and if
the single ACK/NACK information is indicated as NACK information, a
precoding matrix having a functional relationship with a precoding
matrix used in an initial transmission is used as the precoding
matrix for retransmission.
5. The method of claim 2, wherein the PHICH comprises information
about the precoding matrix for retransmission.
6. The method of claim 1, wherein the precoding matrix used in the
retransmission of the at least one transport block is determined
using a sub matrix comprising a column vector used during
transmission of a transport block, decoding of which is failed in
the Node B, among column vectors of a precoding matrix used in an
initial transmission.
7. The method of claim 6, wherein retransmitting the at least one
transport block comprises boosting transmission power when
retransmitting the at least one transport block.
8. The method of claim 1, wherein a precoding matrix in a
predetermined default codebook is used as the precoding matrix used
in retransmission.
9. The method of claim 8, wherein determining the precoding matrix
for retransmission comprises selecting one precoding matrix from
among at least one precoding matrix included in the default
codebook according to a rule determined based on at least one of a
Redundancy Version (RV) value and a system frame number, or the RV
value and a subframe number.
10. The method of claim 1, wherein determining the precoding matrix
for retransmission comprises: if a precoding matrix used in the
initial transmission is defined as a mother matrix, selecting a
precoding matrix to be used in the retransmission from a codebook
comprising at least one child matrix having a predetermined
functional relationship with the mother matrix.
11. The method of claim 1, further comprising: receiving precoding
matrix-related information for the retransmission, transmitted by
the Node B during resource allocation for initial transmission; and
selecting the precoding matrix for the retransmission according to
a rule determined based on at least one of an RV value and a system
frame number or a subframe number.
12. The method of claim 1, wherein determining the precoding matrix
for the retransmission comprises determining, as the precoding
matrix for the retransmission, a sub matrix comprising at least one
column vector in a precoding matrix used in an initial
transmission, according to a rank.
13. The method of claim 1, wherein the precoding matrix for the
retransmission is determined based on channel states of layers over
which transport blocks were transmitted in an initial
transmission.
14. The method of claim 1, wherein determining the precoding matrix
for the retransmission comprises: comparing appropriate Modulation
and Coding Scheme (MCS) levels of transport blocks transmitted in
an initial transmission; and determining, as the precoding matrix
for the retransmission, at least one column vector in a precoding
matrix used in the initial transmission for a transport block
having a higher MCS level.
15. The method of claim 1, wherein determining the precoding matrix
for the retransmission comprises: comparing channel states of
layers over which transport blocks were transmitted in an initial
transmission; and determining, as the precoding matrix for the
retransmission, at least one column vector in a precoding matrix
used in the initial transmission for a transport block transmitted
over a layer having a better channel state.
16. The method of claim 1, wherein determining the precoding matrix
for the retransmission comprises: determining a child codebook from
a precoding matrix used in an initial transmission; and determining
the precoding matrix for the retransmission in the child codebook;
wherein the precoding matrix for the retransmission in the child
codebook is determined in a grant for the initial transmission.
17. The method of claim 1, wherein, if the precoding matrix for the
retransmission is determined from a codebook comprising a plurality
of precoding matrixes, the precoding matrix for the retransmission
is determined using at least one of a unique number of a transport
block to be retransmitted, an RV value, a system frame number, and
a subframe number.
18. The method of claim 1, further comprising: adjusting a
corresponding Physical Uplink Shared CHannel (PUSCH) transmission
for the retransmitting according to the PDCCH and a Physical HARQ
Indicator CHannel (PHICH), if the UE receives the PDCCH and the
PHICH intended for the UE.
19. The method of claim 1, further comprising: adjusting a
Redundancy Version (RV) value in the PDCCH intended for the UE, if
the UE receives the PDCCH intended for the UE, and adjusting a
sequence of the RV to value 0, 2, 3, 1, if the UE does not receive
the PDCCH intended for the UE.
20. The method of claim 1, wherein a number of the NACKed at least
one transport block is less than a number of the plurality of
transport blocks.
21. The method of claim 1, wherein the precoding matrix is
determined as a predefined precoding matrix of the number of at
least one layer corresponding to the at least one transport block
NACKed from the Node B, if the UE does not receive the PDCCH
intended for the UE from the Node B.
22. A User Equipment (UE) for controlling retransmission in a
wireless communication system supporting Multiple Input Multiple
Output (MIMO) technology, comprising: a transceiver for exchanging
data with a Node B over a wireless network; and a controller for
initially transmitting a plurality of transport blocks to the Node
B, receiving a negative acknowledgement (NACK) for at least one
transport block among the plurality of transport blocks, from the
Node B, determining a precoding matrix for retransmission of the at
least one transport block in response to the NACK for the at least
one transport block, and retransmitting the at least one transport
block using the determined precoding matrix, wherein the precoding
matrix is determined based on a number of layers corresponding to
the at least one transport block negatively acknowledged (NACKed)
from the Node B, if the UE does not receive a Physical Downlink
Control CHannel (PDCCH) intended for the UE from the Node B.
23. The UE of claim 22, wherein the controller is further
configured to adjust a corresponding Physical Uplink Shared CHannel
(PUSCH) transmission for the retransmitting according to the PDCCH
and a Physical HARQ Indicator CHannel (PHICH), if the UE receives
the PDCCH and the PHICH intended for the UE.
24. The UE of claim 22, wherein the controller is further
configured to adjust a Redundancy Version (RV) value in the PDCCH
intended for the UE, if the UE receives the PDCCH intended for the
UE, and adjust a sequence of the RV to 0, 2, 3, 1, if the UE does
not receive the PDCCH intended for the UE.
25. The UE of claim 22, wherein a number of the NACKed at least one
transport block is less than a number of the plurality of transport
blocks.
26. The UE of claim 22, wherein the precoding matrix is determined
as a predefined precoding matrix of the number of at least one
layer corresponding to the at least one transport block NACKed from
the Node B, if the UE does not receive the PDCCH intended for the
UE from the Node B.
27. A method for controlling retransmission by a Node B in a
wireless communication system supporting Multiple Input Multiple
Output (MIMO) technology, comprising the steps of: receiving a
plurality of transport blocks transmitted by a User Equipment (UE)
in an initial transmission; transmitting a negative acknowledgement
(NACK) for at least one transport block to the UE, if at least one
transport block among the plurality of transport blocks has failed
to be decoded; determining a precoding matrix that the UE uses
during retransmission of the at least one transport block in
response to the NACK for the at least one transport block; and
receiving the at least one transport block retransmitted using the
determined precoding matrix, wherein the precoding matrix is
determined based on a number of layers corresponding to the at
least one transport block negatively acknowledged (NACKed) from the
Node B, if the UE does not receive a Physical Downlink Control
CHannel (PDCCH) intended for the UE from the Node B.
28. The method of claim 27, further comprising: adjusting reception
of a corresponding Physical Uplink Shared CHannel (PUSCH) for the
retransmission according to the PDCCH and a Physical HARQ Indicator
CHannel (PHICH), if the Node B transmits the PDCCH and the PHICH
intended for the UE.
29. The method of claim 27, further comprising: adjusting a
Redundancy Version (RV) value in the PDCCH intended for the UE, if
the Node B transmits the PDCCH intended for the UE, and receiving a
corresponding PUSCH retransmission adjusted sequence of the RV to
0, 2, 3, 1, if the Node B does not transmit a PDCCH intended for
the UE.
30. The method of claim 27, wherein a number of the NACKed at least
one transport block is less than a number of the plurality of
transport blocks.
31. The method of claim 27, wherein the precoding matrix is
determined as a predefined precoding matrix of the number of at
least one layer corresponding to the at least one transport block
NACKed from the Node B, if the UE does not receive the PDCCH
intended for the UE from the Node B.
32. A Node B for controlling retransmission in a wireless
communication system supporting Multiple Input Multiple Output
(MIMO) technology, comprising: a transceiver for exchanging data
with a User Equipment (UE) over a wireless network; and a
controller for receiving a plurality of transport blocks
transmitted by the UE in an initial transmission, transmitting a
negative acknowledgement (NACK) for at least one transport block to
the UE, if at least one transport block among the plurality of
transport blocks has failed to be decoded, determining a precoding
matrix that the UE uses during retransmission of the at least one
transport block in response to the NACK for the at least one
transport block, and receiving the at least one transport block
retransmitted using the determined precoding matrix, wherein the
precoding matrix is determined based on a number of layers
corresponding to the at least one transport block negatively
acknowledged (NACKed) from the Node B, if the UE does not receive a
Physical Downlink Control CHannel (PDCCH) intended for the UE from
the Node B.
33. The Node B of claim 32, wherein the controller is further
configured to adjust reception of a corresponding Physical Uplink
Shared CHannel (PUSCH) for the retransmission according to the
PDCCH and a Physical HARQ Indicator CHannel (PHICH), if the Node B
transmits the PDCCH and the PHICH intended for the UE.
34. The Node B of claim 32, wherein the controller is further
configured to adjust a Redundancy Version (RV) value in the PDCCH
intended for the UE, if the Node B transmits the PDCCH intended for
the UE, and receive a corresponding PUSCH retransmission adjusted
sequence of the RV to 0, 2, 3, 1, if the Node B does not transmit a
PDCCH intended for the UE.
35. The Node B of claim 32, wherein a number of the NACKed at least
one transport block is less than a number of the plurality of
transport blocks.
36. The Node B of claim 32, wherein the precoding matrix is
determined as a predefined precoding matrix of the number of at
least one layer corresponding to the at least one transport block
NACKed from the Node B, if the UE does not receive the PDCCH
intended for the UE from the Node B.
Description
PRIORITY
This application claims priority under 35 U.S.C. .sctn.119(a) to
Korean Patent Applications filed in the Korean Intellectual
Property Office on Mar. 29, 2010 and Apr. 19, 2010 and assigned
Serial Nos. 10-2010-28207 and 10-2010-36134, respectively, the
entire disclosures of which are incorporated herein by
reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a method and apparatus
for retransmission on an UpLink (UL) in a wireless communication
system, and more particularly, to a method and apparatus for
controlling retransmission on a UL in a wireless communication
system supporting multi-antenna transmission technology, such as
Multiple Input Multiple Output (MIMO).
2. Description of the Related Art
Wireless communication systems have evolved into broadband wireless
communication systems providing not only voice-oriented services
but also high-speed, high-quality packet data services, including
communication standards such as, for example, 3GPP High Speed
Packet Access (HSPA), Long Term Evolution (LTE), 3GPP2 High Rate
Packet Data (HRPD), Ultra Mobile Broadband (UMB), and IEEE
802.16e.
Recently, in order to improve transmission efficiency, wireless
communication systems use technologies such as Adaptive Modulation
and Coding (AMC) and channel-sensitive scheduling. By using AMC, a
Node B (also known as a base station) may adjust the amount of data
transmitted by the Node B or a User Equipment (UE), also known as a
mobile station, according to channel states. For example, if the
channel state is poor, the amount of transmission data is reduced
to a desired level to match a reception error rate, and if the
channel state is good, the amount of transmission data is increased
to effectively transmit as much information as possible, while
matching the reception error rate to the desired level. By using a
channel-sensitive scheduling resource management method, a Node B
may selectively service users with a good channel state among a
plurality of users, which contributes to an increase in system
capacity, as compared to an existing method of allocating a channel
to a single user and servicing the user. Specifically, the AMC and
the channel-sensitive scheduling are methods of applying an
appropriate Modulation and Coding Scheme (MCS) at the time
determined to be most efficient, using the channel state
information.
Many studies are being conducted to replace Code Division Multiple
Access (CDMA), a multiple access scheme which has been used in the
2.sup.nd Generation (2G) and 3.sup.rd Generation (3G) mobile
communication systems, with Orthogonal Frequency Division Multiple
Access (OFDMA) in the next-generation communication system.
Standards bodies such as 3GPP, 3GPP2, and IEEE are standardizing
evolved systems using OFDMA or modified OFDMA. It is well known
that a greater capacity increase can be expected in OFDMA, compared
to in CDMA. One of several reasons leading to this capacity
increase in OFDMA is the possibility of performing scheduling on
the frequency domain (known as `Frequency-Domain Scheduling`). Just
as the capacity gain can be obtained by the channel-sensitive
scheduling using the time-varying characteristics of channels, a
higher capacity gain may be obtained using the frequency-varying
characteristics of channels.
An LTE system, a typical example of the broadband wireless
communication systems, adopts Orthogonal Frequency Division
Multiplexing (OFDM) in a Downlink (DL) and Single Carrier Frequency
Division Multiple Access (SC-FDMA) in a UL, both of which may
perform the frequency-domain scheduling.
The AMC and channel-sensitive scheduling are techniques capable of
improving the transmission efficiency when transmitters have
acquired sufficient information about transmission channels. In the
DL of the LTE system, for Frequency Division Duplex (FDD), since a
Node B cannot estimate a state of a DL channel or a transmission
channel depending on a UL channel or a reception channel, a UE is
designed to report information about the DL channel to the Node B.
In the case of Time Division Duplex (TDD), a Node B uses
characteristics so that it can estimate a state of a DL channel
depending on a UL channel, making it possible to omit the process
of reporting the information about the DL channel from the UE to
the Node B.
In the UL of the LTE system, a UE is designed to transmit a
Sounding Reference Signal (SRS) and a Node B is designed to
estimate a UL channel by receiving the SRS.
In the DL of the LTE system, MIMO or a multi-antenna transmission
technique is supported. An LTE Node B may include 1, 2, or 4
transmission antennas. When including a plurality of transmission
antennas, a Node B may obtain a beamforming gain and a spatial
multiplexing gain by applying precoding.
Recently, many discussions have been held in 3GPP to support MIMO
even in the UL of the LTE system. Similar to the DL MIMO, a UE may
include 1, 2, or 4 transmission antennas, and when including a
plurality of transmission antennas, a UE may obtain a beamforming
gain and a spatial multiplexing gain by applying precoding.
A difference between the DL MIMO and the UL MIMO is provided below.
In the DL MIMO, a Node B (or a transmitter) determines by itself
the transmission property such as MCS scheme, MIMO scheme, and
precoding. The Node B configures and transmits a Physical Downlink
Shared CHannel (PDSCH) by reflecting the transmission property, and
delivers the transmission property applied to the PDSCH to a UE
using a Physical Downlink Control CHannel (PDCCH). However, in the
UL MIMO, a Node B (or a receiver) determines the transmission
property such as MCS scheme, MIMO scheme, and precoding, according
to the channel characteristics of UEs. The Node B delivers the
transmission property to a UE through a PDCCH, and upon receiving
the PDCCH, the UE configures and transmits a Physical Uplink Shared
CHannel (PUSCH) by reflecting the transmission property granted by
the Node B. Specifically, in the LTE system, a Node B determines
AMC, channel-sensitive scheduling, MIMO precoding, etc., and a UE
receives a PDSCH transmitted based on the determination, or
configures and transmits a PUSCH according to the
determination.
If a Node B has correct information about a channel state, the Node
B may determine an amount of transmission data, which is most
appropriate for the channel state, using AMC. In actual
communication environments, however, there is a significant
difference between the channel state that the Node B is aware of,
and the actual channel state, due to the estimation error, the
feedback error, and the like. Therefore, despite the use of AMC,
the transmitter and the receiver may not actually prevent errors
from occurring. The majority of wireless communication systems,
including the LTE system, employ Hybrid Automatic ReQuest (HARQ),
in which, if a decoding failure occurs in an initial transmission,
a physical layer immediately retransmits the failed data. HARQ
refers to a scheme, in which, if a receiver has failed to correctly
decode data, the receiver transmits NACK information indicating the
decoding failure to a transmitter, allowing the transmitter to
retransmit the failed data in a physical layer. On the contrary, if
the receiver has correctly decoded data, the receiver transmits ACK
information indicating the decoding success to the transmitter,
allowing the transmitter to transmit new data.
In a wireless communication system using the HARQ, because a
receiver may improve its reception performance by combining a
retransmitted signal with a previously received signal, the
receiver stores in its memory the data which was received
previously but failed to be decoded, just in case of
retransmission.
In order to enable a transmitter to transmit other data for the
time required when a response signal from a receiver, such as ACK
and NACK, is delivered up to the transmitter, an HARQ process is
defined. In accordance with the HARQ process, the receiver may
determine whether to combine a previously received signal with a
newly received signal using a HARQ Process Identification (HARQ
PID). HARQ is classified into synchronous HARQ and asynchronous
HARQ according to whether a transmitter provides the HARQ PID to a
receiver as a control signal in the HARQ process. In the
synchronous HARQ, a transmitter uses a serial number of a subframe
carrying a PDCCH, instead of providing a HARQ PID to a receiver as
a control signal. The subframe refers to a resource allocation unit
in the time domain. However, in the asynchronous HARQ, a
transmitter provides a HARQ PID to a receiver as a control signal.
The LTE system employs asynchronous HARQ in the DL and synchronous
HARQ in the UL.
FIG. 1 illustrates a synchronous HARQ operation in a UL.
Referring to FIG. 1, if a Node B grants a resource allocation for a
UL transmission using a PDCCH in an n-th subframe of a DL in step
101, a HARQ PID is determined as resource allocation information by
a subframe serial number `n`. For example, if a HARQ PID
corresponding to a subframe serial number `n` is assumed to be `0`,
a HARQ PID corresponding to a subframe serial number `n+1` may be
defined as `1`. A PDCCH for a UL grant, transmitted in a subframe
with a serial number `n`, includes a New Data Indicator (NDI). If
an NDI has been toggled from its previous NDI value, the relevant
UL grant is set to allocate a PUSCH for new data transmission. If
an NDI has maintained its previous NDI value, the relevant UL grant
is set to allocate a PUSCH for retransmission of the previously
transmitted data.
If an NDI associated a UL grant is assumed to be toggled in step
101, a UE performs initial transmission on a PUSCH for new data
transmission in a subframe #(n+4) in step 103. Whether the Node B
has successfully decoded the PUSCH data transmitted by the UE in
the subframe #(n+4) is determined using a Physical HARQ Indicator
CHannel (PHICH) that the Node B transmits in a subframe #(n+8) in
step 105. If it is determined that the PHICH has transmitted a
NACK, the UE performs retransmission on the PUSCH in a subframe
#(n+12) in step 107. In this way, in the synchronous HARQ, initial
transmission and retransmission of the same Transport Block (TB)
are performed in sync with serial numbers of subframes.
As described in FIG. 1, the Node B and the UE may normally perform
a HARQ operation without introducing a separate HARQ PID, because
an agreement was made in advance that the TB having undergone
initial transmission in the subframe #(n+4) is retransmitted in the
subframe #(n+12). In the example of FIG. 1, since a transmission
interval of the same TB includes 8 subframes, the maximum number of
HARQ processes capable of running at the same time may be limited
to 8.
In the UL synchronous HARQ operation described in FIG. 1,
retransmission may be granted using a PHICH capable of indicating
only the ACK/NACK signal. If the Node B desires to change the
transmission property of a PUSCH, such as a transmission resource
and an MCS scheme, in retransmission, the Node B may grant
transmission of a PDCCH indicating the change. This HARQ scheme,
granting a change in the transmission property of the PUSCH, is
called `adaptive synchronous HARQ`.
FIG. 2 illustrates an adaptive synchronous HARQ operation in a
UL.
Referring to FIG. 2, steps 101 to 105 in FIG. 2 are identical in
operation to their corresponding steps in FIG. 1.
In step 105 in FIG. 2, a Node B informs a UE that it has failed to
successfully decode the PUSCH transmitted in the subframe #(n+4) in
step 103, by delivering a NACK using a PHICH in a subframe #(n+8).
In order to change the transmission property during PUSCH
retransmission, the Node B transmits a PDCCH including information
for changing the transmission property of a PUSCH, together with
the PHICH in step 106. The UE may receive the PDCCH including
information for changing the transmission property of a PUSCH,
because it attempts to receive and decode a PDCCH in every
subframe. In step 108, the UE performs retransmission on a PUSCH in
a subframe #(n+12) by applying the transmission property indicated
by the PDCCH.
According to the above-described adaptive synchronous HARQ, the
information for changing the transmission property of a PUSCH is
transmitted over a PDCCH. Therefore, if a change in the
transmission property of a PUSCH is required during retransmission,
a Node B transmits a PDCCH together with a PHICH despite an
increase in the amount of DL control information. When maintaining
the previous transmission property of a PUSCH, the Node B transmits
only the PHICH.
FIG. 3 illustrates an adaptive synchronous HARQ operation of a Node
B in a UL.
Referring to FIG. 3, in step 131, a Node B performs UL scheduling
to determine a UE to be granted transmission of a PUSCH, and a
resource to be used for the PUSCH transmission. In step 133, the
Node B transmits a PDCCH to inform the scheduled UE of grant
information of the PUSCH. In step 135, the Node B demodulates and
decodes the PUSCH, which has been received four subframes after a
time when the PDCCH was transmitted in step 133. In step 137, the
Node B determines whether the decoding of the PUSCH is successful.
If successful, the Node B transmits an ACK to the UE in step 139,
and then returns to step 131 to perform new scheduling. On the
other hand, if the decoding is failed in step 137, the Node B
transmits a NACK to the UE in step 141.
Thereafter, in accordance with an adaptive synchronous HARQ
operation, the Node B determines in step 143 whether it desires to
change the transmission property of the PUSCH to be different from
that designated in step 133. If it is desired to change the
transmission property, the Node B transmits a PDCCH including
information indicating a new transmission property to be applied
for retransmission of the PUSCH in step 145. After indicating
retransmission of the PUSCH in steps 143 and 145, the Node B
returns to step 135 to receive and decode the retransmitted
PUSCH.
FIG. 4 illustrates an adaptive synchronous HARQ operation of a UE
in a UL.
Referring to FIG. 4, a UE attempts to receive and decode a PDCCH
for a UL grant in step 151, and determines in step 153 whether the
decoding of the PDCCH is successful. If successful, the UE
determines in step 155 whether an NDI indicating
transmission/non-transmission of new data has been toggled. If the
NDI has been toggled, meaning that the relevant grant indicates
initial transmission of a new TB, then the UE transmits a PUSCH
carrying a new TB by applying the transmission property indicated
by the PDCCH in step 157. However, if the NDI has not been toggled
in step 155, meaning that the relevant grant indicates
retransmission with the transmission property changed because a
Node B has failed to successfully decode the previous TB having the
same HARQ PID, then the UE retransmits a PUSCH carrying the
previous TB by applying the transmission property indicated by the
PDCCH in step 159. If the UE has failed to successfully decode the
PDCCH for a UL grant in step 153, the UE attempts to receive and
decode a PHICH in step 161. In step 163, the UE determines if an
ACK has been received over the PHICH. Upon receiving the ACK, the
UE stops the transmission of the PUSCH in step 165. However, upon
receiving a NACK from the PHICH, the UE retransmits a PUSCH
carrying the previous TB by applying the transmission property
indicated by the last received PDCCH in step 167.
Although the synchronous HARQ has been proposed to enable
retransmission by a UE by transmitting only the PHICH without
transmitting the PDCCH, when the PDCCH should be transmitted
together with the PHICH to indicate the transmission property such
as a precoding scheme of a UE, the above resource saving effects
may not be expected in the synchronous HARQ. Specifically, while
the PHICH carries only the ACK/NACK information, the PDCCH includes
various control information for UL transmission in a UE. Therefore,
to transmit the PDCCH, a Node B should consume more frequency
resources and transmission power. If the PDCCH is to be transmitted
to indicate the transmission property, such as a precoding scheme
for MIMO transmission, during retransmission in a UL, the
consumption of resources for control information increases,
requiring a method for reducing a transmission load of control
information for retransmission in a UL.
SUMMARY OF THE INVENTION
The present invention has been made to address at least the above
problems and/or disadvantages and to provide at least the
advantages described below. Accordingly, an aspect of the present
invention is provides a method and apparatus for efficiently
controlling retransmission on a UL in a wireless communication
system supporting MIMO.
Another aspect of the present invention provides a retransmission
control method and apparatus capable of reducing transmission of
control information for retransmission on a UL in a wireless
communication system supporting UL MIMO.
Further, another aspect of the present invention provides a
retransmission control method and apparatus capable of reducing
transmission of control information indicating the transmission
property during retransmission in a wireless communication system
supporting UL MIMO.
Yet another aspect of the present invention provides a method and
apparatus for efficiently determining a precoding scheme during
retransmission on a UL in a wireless communication system
supporting UL MIMO.
Still another aspect of the present invention provides a method and
apparatus for determining a precoding scheme during retransmission
taking into account channel states of layers over which TBs, having
undergone initial transmission, are transmitted in a UL, in a
wireless communication system supporting UL MIMO.
According to one aspect of the present invention, a method is
provided for controlling retransmission by a UE in a wireless
communication system supporting MIMO technology. A plurality of
transport blocks are initially transmitted to a Node B, and a
retransmission request for at least one transport block among the
plurality of transport blocks is received from the Node B. A
precoding matrix for retransmission of the at least one transport
block is determined based on the retransmission request for the at
least one transport block. The at least one transport block is
retransmitted using the determined precoding matrix.
According to another aspect of the present invention, a UE is
provided for controlling retransmission in a wireless communication
system supporting MIMO technology. The UE includes a transceiver
for exchanging data with a Node B over a wireless network. The UE
also includes a controller for initially transmitting a plurality
of transport blocks to the Node B, receiving a retransmission
request for at least one transport block among the plurality of
transport blocks, from the Node B, determining a precoding matrix
for retransmission of the at least one transport block based on the
retransmission request for the at least one transport block, and
retransmitting the at least one transport block using the
determined precoding matrix.
According to an additional aspect of the present invention, a
method is provided for controlling retransmission by a Node B in a
wireless communication system supporting MIMO technology. A
plurality of transport blocks transmitted by a UE in an initial
transmission are received. A retransmission request is transmitted
to the UE when at least one transport block among the plurality of
transport blocks has failed to be decoded. A precoding matrix that
the UE uses during retransmission of the at least one transport
block is determined based on the retransmission request for the at
least one transport block. The at least one transport block
retransmitted using the determined precoding matrix is
received.
According to a further aspect of the present invention, a Node B is
provided for controlling retransmission in a wireless communication
system supporting MIMO technology. The Node B includes a
transceiver for exchanging data with a UE over a wireless network.
The Node B also includes a controller for receiving a plurality of
transport blocks transmitted by the UE in an initial transmission,
transmitting a retransmission request to the UE when at least one
transport block among the plurality of transport blocks has failed
to be decoded, determining a precoding matrix that the UE uses
during retransmission of the at least one transport block, based on
the retransmission request for the at least one transport block,
and receiving the at least one transport block retransmitted using
the determined precoding matrix.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other aspects, features and advantages of the present
invention will be more apparent from the following description when
taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a diagram illustrating a synchronous HARQ operation in a
UL;
FIG. 2 is a diagram illustrating an adaptive synchronous HARQ
operation in a UL;
FIG. 3 is a flowchart illustrating an adaptive synchronous HARQ
operation of a Node B in a UL;
FIG. 4 is a flowchart illustrating an adaptive synchronous HARQ
operation of a UE in a UL;
FIG. 5 is a block diagram illustrating a structure of a UE in a
wireless communication system supporting UL MIMO, according to an
embodiment of the present invention;
FIG. 6 is a block diagram illustrating a structure of a Node B in a
wireless communication system supporting UL MIMO, according to an
embodiment of the present invention;
FIG. 7 is a flowchart illustrating a general operation of a Node B
in a wireless communication system supporting UL MIMO;
FIG. 8 is a flowchart illustrating a general operation of a UE in a
wireless communication system supporting UL MIMO;
FIG. 9 is a flowchart illustrating a procedure for determining a
precoding scheme in a Node B supporting UL MIMO, according to an
embodiment of the present invention;
FIG. 10 is a flowchart illustrating a procedure for determining a
precoding scheme in a UE supporting UL MIMO, according to an
embodiment of the present invention;
FIG. 11 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a first
embodiment of the present invention;
FIG. 12 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the first
embodiment of the present invention;
FIG. 13 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a second
embodiment of the present invention;
FIG. 14 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the
second embodiment of the present invention;
FIG. 15 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a third
embodiment of the present invention;
FIG. 16 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the third
embodiment of the present invention;
FIG. 17 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a fourth
embodiment of the present invention;
FIG. 18 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the
fourth embodiment of the present invention;
FIG. 19 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a fifth
embodiment of the present invention;
FIG. 20 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the fifth
embodiment of the present invention;
FIG. 21 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a sixth
embodiment of the present invention;
FIG. 22 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the sixth
embodiment of the present invention;
FIG. 23 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to an eighth
embodiment of the present invention;
FIG. 24 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the
eighth embodiment of the present invention;
FIG. 25 is a flowchart illustrating a precoding determining method
performed in a UE during retransmission, according to a ninth
embodiment of the present invention; and
FIG. 26 is a flowchart illustrating a precoding determining method
performed in a Node B during retransmission, according to the ninth
embodiment of the present invention.
DETAILED DESCRIPTION OF EMBODIMENTS OF THE PRESENT INVENTION
Embodiments of the present invention are described in detail with
reference to the accompanying drawings. The same or similar
components may be designated by the same or similar reference
numerals although they are illustrated in different drawings. In
the following description, specific details such as detailed
configurations and components are merely provided to assist the
overall understanding of embodiments of the present invention.
Therefore, it should be apparent to those skilled in the art that
various changes and modifications of the embodiments described
herein can be made without departing from the scope and spirit of
the invention. In addition, detailed descriptions of constructions
or processes known in the art may be omitted to avoid obscuring the
subject matter of the present invention.
Although an LTE system will be considered in the following
description of embodiments of the present invention, the present
invention may be applied in the same way not only to the LTE
system, but also to any wireless communication systems that support
UL MIMO and provide, to UEs, information about the changed
transmission property through a control channel if there is a need
for a change in the transmission property used during
retransmission, such as a precoding scheme.
A UL of the LTE system, to which embodiments of the present
invention is applied, adopts SC-FDMA. In this regard, Uplink
Control Information (UCI) including UL ACK/NACK information for
HARQ in a DL and feedback information such as Channel Quality
Indicator (CQI), Precoding Matrix Indicator (PMI), and Rank
Indicator (RI), is transmitted on a Physical Uplink Control CHannel
(PUCCH), and UL data is transmitted on a PUSCH.
In order to maintain the single-carrier property in the SC-FDMA,
when the UCI and the UL data are to be transmitted simultaneously,
the UCI is multiplexed with the data signal in the PUSCH without
being transmitted on the PUCCH. If an aperiodic CQI is requested as
a UL grant, UCI and data should be multiplexed because not only the
data, but also the aperiodic CQI, PMI, and RI, should be included
in the PUSCH before transmission.
FIG. 5 illustrates a structure of a UE in a wireless communication
system supporting UL MIMO according to an embodiment of the present
invention, in which a transmitter structure of the UE is shown.
Referring to FIG. 5, a block 201 codes and modulates an input data
signal, and a block 205 codes and modulates an input UCI signal. In
a UE supporting UL MIMO, a maximum of two CodeWords (CWs) are
created as represented by reference numeral 203. Generally, CWs
correspond to TBs. Assuming that CW0 is identical to TB1 and CW1 is
identical to TB2, if a swap function is enabled, then a mapping
relationship between the CWs and the TBs may be changed. After the
change, the CW0 is mapped to the TB2, and the CW1 is mapped to the
TB1. While the swap function is defined in DL MIMO of the LTE
system, the swap function is optional in UL MIMO.
Among the lines represented by reference numeral 203, a solid line
indicates that one CW is created, and a dotted line indicates that
two CWs may be created. A data modulation signal made in the block
201 and a UCI modulation signal made in the block 205 undergo
multiplexing and interleaving, and then are mapped to MIMO layers
in block 207. An example of a method of mapping CWs to layers in
the LTE system is shown below in Table 1.
TABLE-US-00001 TABLE 1 CW-to-Layer Mapping Number of Number of
Layers (rank) CWs CW-to-Layer Mapping 1 1 CW 0 .fwdarw. layer 0:
x.sup.(0)(i) = d.sup.(0)(i) 2 2 CW 0 .fwdarw. layer 0 & CW 1
.fwdarw. layer 1: x.sup.(0)(i) = d.sup.(0)(i) x.sup.(1)(i) =
d.sup.(1)(i) 2 1 CW 0 .fwdarw. layers {0, 1}: x.sup.(0)(i) =
d.sup.(0)(2i) x.sup.(1)(i) = d.sup.(0)(2i + 1) allowed only in
retransmission 3 2 CW 0 .fwdarw. layer 0 & CW 1 .fwdarw. layers
{1, 2}: x.sup.(0)(i) = d.sup.(0)(i) x.sup.(1)(i) = d.sup.(1)(2i)
x.sup.(2)(i) = d.sup.(1)(2i + 1) 4 2 CW 0 .fwdarw. layers {0, 1}
& CW 1 .fwdarw. layers {2, 3}: x.sup.(0)(i) = d.sup.(0)(i)
x.sup.(1)(i) = d.sup.(1)(2i) x.sup.(2)(i) = d.sup.(1)(2i + 1)
In Table 1, d.sup.(k)(i) represents an i-th modulation symbol in a
CWk, and x.sup.(l)(i) represents an i-th symbol on an l-th layer.
When one CW is mapped to two layers, even-numbered modulation
symbols are mapped to a lower layer, while odd-numbered modulation
symbols are mapped to an upper layer. More modulation symbols may
be transmitted, contributing to an increase in the amount of
transmission data and a reduction in coding rate, compared to when
one CW is mapped to one layer.
As shown in Table 1, for a rank-1 transmission in which only one
layer is made, only one CW is created, and for a rank-1
transmission in which two layers are made, two CWs are created.
There is also a case where one CW is created in a rank-2
transmission, and this case is allowed only in retransmission.
Generally, with respect to a relationship between ranks and layers
in MIMO, the term `layer` refers to a spatial resource capable of
transmitting one modulation symbol stream, and the `rank` refers to
the number of layers formed in the MIMO system. A MIMO-based
spatial multiplexing technique increases a data rate by making
multiple layers for the same time-frequency resource and
transmitting independent modulation symbol streams on the
layers.
The layer-specific signals generated by the block 207 undergo
transmission (Tx) precoding in a block 209a. Precoding is a process
of forming beams of layers to increase reception quality of the
layers. Precoding should be determined taking into account
characteristics of transmission channels. With respect to UL MIMO,
since a transmission channel is a UL channel, if a Node B measures
the UL channel and informs a UE of a precoder employing an
appropriate precoding scheme, the UE performs precoding according
to the information. A precoder is represented in a matrix (i.e.,
precoding matrix), in which the number of rows is equal to the
number of antennas and the number of columns is equal to the number
of layers. Precoding may be expressed in general formula as shown
in Equation (1) below.
.function..function..times..times.
.times..times..function..function..function. ##EQU00001## P
represents a precoding matrix, x represents a transmission signal
before undergoing precoding, y represents a transmission signal
after undergoing precoding, and x.sup.(n)(i) represents an i-th
symbol to be transmitted via an n-th transmission antenna. For
reference, the term `transmission antenna` as used herein refers to
a logical antenna used for signal transmission, rather than a
physical antenna. Mapping between logical antennas and physical
antennas may be defined in many different ways.
Tables 2 and 3 below show examples of precoding matrices used for
LTE UL MIMO in different scenarios where two and four transmission
antennas are used, respectively.
TABLE-US-00002 TABLE 2 Example of Precoding Matrix for LTE UL MIMO
(with 2 Tx Antennas) Rank- 1 Indices 0~3 .function. ##EQU00002##
.function. ##EQU00003## .function. ##EQU00004## .function.
##EQU00005## Indices 4~5 .function. ##EQU00006## .function.
##EQU00007## -- -- Rank- 2 Index 0 .function. ##EQU00008## -- --
--
TABLE-US-00003 TABLE 3 Example of Precoding Matrix for LTE UL MIMO
(with 4 Tx Antennas) Rank-1 Indices 0~3 .function. ##EQU00009##
.function. ##EQU00010## .function. ##EQU00011## .function.
##EQU00012## Indices 4~7 .function. ##EQU00013## .function.
##EQU00014## .function. ##EQU00015## .function. ##EQU00016##
Indices 8~11 .function. ##EQU00017## .function. ##EQU00018##
.function. ##EQU00019## .function. ##EQU00020## Indices 12~15
.function. ##EQU00021## .function. ##EQU00022## .function.
##EQU00023## .function. ##EQU00024## Indices 15~19 .function.
##EQU00025## .function. ##EQU00026## .function. ##EQU00027##
.function. ##EQU00028## Indices 20~23 .function. ##EQU00029##
.function. ##EQU00030## .function. ##EQU00031## .function.
##EQU00032## Rank-2 Indices 0~3 .function. ##EQU00033## .function.
##EQU00034## .function. ##EQU00035## .function. ##EQU00036##
Indices 4~7 .function. ##EQU00037## .function. ##EQU00038##
.function. ##EQU00039## .function. ##EQU00040## Indices 8~11
.function. ##EQU00041## .function. ##EQU00042## .function.
##EQU00043## .function. ##EQU00044## Indices 12~15 .function.
##EQU00045## .function. ##EQU00046## .function. ##EQU00047##
.function. ##EQU00048## Rank-3 Indices 0~3 .function. ##EQU00049##
.function. ##EQU00050## .function. ##EQU00051## .function.
##EQU00052## Indices 4~7 .function. ##EQU00053## .function.
##EQU00054## .function. ##EQU00055## .function. ##EQU00056##
Indices 8~11 .function. ##EQU00057## .function. ##EQU00058##
.function. ##EQU00059## .function. ##EQU00060## Rank-4 Index 0
.function. ##EQU00061## -- -- --
Referring again to FIG. 5, signals having passed through block 209a
are signals to be transmitted via transmission antennas 217a, . . .
, 217b. These signals are converted into SC-FDMA signals to be
compatible with the LTE UL scheme by means of blocks 211 (211a, . .
. , 211b). The block 211a is an SC-FDMA signal converter for a
signal to be transmitted via the first transmission antenna 217a.
The block 211b is an SC-FDMA signal converter for a signal to be
transmitted via the last transmission antenna 217b. The SC-FDMA
signal converter 211, as illustrated in FIG. 5, includes a Discrete
Fourier Transform (DFT) unit 221, a resource mapper 223, an Inverse
Fast Fourier Transform (IFFT) unit 225, and a Cyclic Prefix (CP)
adder 227, which are well known in the art.
A Reference Signal (RS) is a signal provided for coherent
demodulation. Independent RSs are generated in different layers,
and blocks 231 (231a, . . . , 231b) are RS generators for their
associated layers. The block 231a is an RS generator for the first
layer, and the block 231b is an RS generator for the last layer. In
a block 209b, the RSs of the different layers undergo the same
precoding as that applied to a PUSCH in block 209a. Since the same
precoding is applied to the RSs and the PUSCH, a Node B may receive
the RSs and estimate a channel for demodulating the layer-specific
signals. By applying precoding to the layer-specific RSs, RS
signals to be transmitted via the transmission antennas 217a, . . .
, 217b may be obtained.
SC-FDMA signals in a PUSCH, to be transmitted via the transmission
antennas 217a, . . . , 217b, are multiplexed with the RSs to be
transmitted via the transmission antennas 217a, . . . , 217b by
means of blocks 213 (213a, . . . , 213b). The block 213a is a
PUSCH/RS multiplexer for a signal to be transmitted via the first
transmission antenna 217a, and the block 213b is a PUSCH/RS
multiplexer for a signal to be transmitted via the last
transmission antenna 217b. To maintain the single-carrier property,
RSs are PUSCH undergo time-division multiplexing so as to be
transmitted with different SC-FDMA symbols.
Baseband signals that the UE will transmit via the transmission
antennas 217a, . . . , 217b are converted into Radio Frequency (RF)
signals by means of RF processors 215 (215a, . . . , 215b), and
then transmitted via the transmission antennas 217a, . . . , 217b.
The blocks 215a and 215b are RF processors for signals to be
transmitted via the first and last transmission antennas 217a and
217b, respectively. Reference numerals 217a and 217b represent the
first and last transmission antennas, respectively.
A block 241 is a controller for controlling the overall control
operation of the UE. The controller 241 determines a frequency
resource for transmitting a PUSCH, an MCS level for data and UCI to
be transmitted in a PUSCH, an amount of resources to be allocated
to UCI among the PUSCH resources, a rank for MIMO transmission, a
precoding scheme, and a parameter for generation of
antenna-specific RS signals. The controller 241 controls the
resource mapper 223, the data coding and modulation block 201, the
UCI coding and modulation block 205, the block 207 for performing
multiplexing, interleaving and CW-to-layer mapping on the data and
UCI, the precoders 209a and 209b, and the RS generators 231.
During UL retransmission, the controller 241 determines the
transmission property for PUSCH transmission according to the
schemes determined in all embodiments including first to tenth
embodiments of the present invention, and controls TBs subjected to
retransmission to be retransmitted through the PUSCH. The
transmission property includes a precoding scheme. Embodiments of
the present invention are described in greater detailed below.
FIG. 6 illustrates a structure of a Node B in a wireless
communication system supporting UL MIMO, according to an embodiment
of the present invention, in which a receiver structure of the Node
B is shown.
Referring to FIG. 6, reference numeral 301a represents a first
reception antenna of a Node B, and reference numeral 301b
represents a last reception antenna of the Node B. Signals received
via the multiple reception antennas 301a, . . . , 301b are
converted into baseband signals by means of RF processors 303a, . .
. , 303b, respectively. The block 303a is an RF processor for
processing a signal received via the first reception antenna 301a,
and the block 303b is an RF processor for processing a signal
received via the last reception antenna 301b. The signals received
via their associated reception antennas and converted into baseband
signals are restored to modulation symbol streams in SC-FDMA
receivers 305a, . . . , 305b. The block 305a is an SC-FDMA receiver
for processing a signal received via the first reception antenna
301a, and the block 305b is an SC-FDMA receiver for processing a
signal received via the last reception antenna 301b.
Each of the SC-FDMA receivers 305a, . . . , 305b, as illustrated in
FIG. 6, includes a CP remover 331, a Fast Fourier Transform (FFT)
unit 333, a resource demapper 335, and an Inverse Discrete Fourier
Transform (IDFT) unit 337, and inversely performs the process of
the SC-FDMA signal converters 211a, . . . , 211b in FIG. 5.
Signals having passed through the SC-FDMA receivers 305a, . . . ,
305b are received signals of PUSCH and RSs from a specific UE.
Because the PUSCH and RSs have undergone time-division
multiplexing, they are separated into PUSCH received signals and RS
received signals by demultiplexers 307a, . . . , 307b. The block
307a is a demultiplexer for processing a signal received via the
first reception antenna 301a, and the block 307b is a demultiplexer
for processing a signal received via the last reception antenna
301b. The RS received signals extracted through the demultiplexing
process are delivered to a channel estimator 311. The PUSCH
received signals extracted through the demultiplexing process are
delivered to a MIMO reception filter 315.
The channel estimator 311 estimates a UL channel from the RS
received signals, and delivers the channel estimate to a controller
313 so that the controller 313 may calculate appropriate reception
filter coefficients. The reception filter coefficients determined
by the controller 313 are delivered to the MIMO reception filter
315. The MIMO reception filter 315 inversely performs the
operations of the precoders 209 in FIG. 5, and separates
layer-specific signals of a PUSCH. Typically, the MIMO reception
filter may include a Minimum Mean Square Error (MMSE) reception
filter. Various other known reception filters may also be used.
The layer-specific received signals are converted into a
CW-specific modulation signal stream 319 and a UCI modulation
signal stream by a block 317, which inversely performs the
operation of the block 207 in FIG. 5. Specifically, the block 317
performs a process of gathering layer-specific signals back on a CW
basis, and a series of deinterleaving and data/UCI demultiplexing
processes. This series of processes are handled according to the
control information, which was transmitted from the Node B to a UE
in advance, under control of the controller 313.
The CW-specific modulation signal stream 319 output from block 317
is demodulated and decoded into the original data by a block 321.
The UCI modulation signal stream output from block 317 is
demodulated and decoded into an original UCI signal by a block 323.
After undergoing the data/UCI decoding process, the decoded data
and UCI are delivered to the controller 313, in order to enable the
Node B to perform UL/DL scheduling and AMC according to the
success/failure in data reception, and the UCI information.
During UL retransmission, the controller 313 determines the
transmission property for PUSCH transmission according to the
schemes determined in all embodiments of the present invention, and
controls the overall reception operation of TBs retransmitted
through the PUSCH. The transmission property includes a precoding
scheme. Embodiments of the present invention are described in
greater detailed below.
Prior to a detailed description of the embodiments of the present
invention, a general operation of transmitting and receiving a
PHICH indicating ACK/NACK information and a PDCCH indicating the
transmission property of a PUSCH in a Node B and UE during
retransmission in UL MIMO will be described with reference to FIGS.
7 and 8. As an example of the transmission property, a precoding
scheme will be considered herein.
Regarding the general information configuration of grant control
information or Downlink Control Information (DCI) in a PDCCH, by
which a Node B instructs a UE to transmit a PUSCH for UL MIMO, the
DCI includes the following Information Element (IEs).
1) Identification flag for a DCI format 0 and a DCI format 1A: In
the LTE system, as DCI is defined to have sizes of a DCI format 0
for a UL grant and a DCI format 1A for compact DL allocation, an IE
by which it can be determined whether the DCI is for a format 0 or
a format 1A is required. This flag is used for that purpose.
2) Frequency hopping flag: This flag is an IE indicating whether
frequency hopping is applied for transmission of a PUSCH to obtain
frequency diversity.
3) Resource allocation information: This resource allocation
information IE is defined to indicate a frequency resource by which
a PUSCH should be transmitted.
4) MCS level: This is an IE indicating an MCS level used for PUSCH
transmission. Some codepoints in this IE are defined to designate a
Redundancy Version (RV) in retransmission.
5) NDI: This is an IE indicating whether a relevant grant is for
transmission of a new TB, or for retransmission of a TB. If there
is a change in the previous value of an NDI, the IE is a grant for
transmission of a new TB. If there is no change, the IE is a grant
for retransmission.
6) Power control information: This is an IE indicating information
about the transmission power used for PUSCH transmission.
7) RS parameter: An RS for PUSCH demodulation is defined as a
Zadoff-Chu (ZC) sequence. The ZC sequence is characterized to
become a new orthogonal ZC sequence if a cyclic shift is changed.
An IE indicating a cyclic shift of an RS for PUSCH demodulation is
defined in a UL grant, for multi-user UL MIMO. If RSs with
different cyclic shifts are allocated to different users, a Node B
may distinguish different user signals using the orthogonality
among RSs.
8) Channel Quality Indicator (CQI) request: This is an IE for
enabling transmission of an aperiodic CQI feedback with a PUSCH.
This IE is defined with 1 bit. If its value is 1, not only the
data, but also an aperiodic CQI, a PMI, and an RI are transmitted
in a PUSCH. If its value is 0, only data is transmitted in a
PUSCH.
In addition, the following IEs are defined in control information
DCI of a grant, by which a Node B instructs a UE to transmit a
PUSCH for UL MIMO.
1) PMI: This is an IE indicating a precoding scheme as the
transmission property used for UL MIMO transmission.
2) MCS level for a second TB: A maximum of two TBs can be
transmitted by UL MIMO. Therefore, an IE indicating an MCS level
for the second TB is defined.
3) NDI for a second TB: For UL MIMO, independent NDIs may be
defined for two TBs, or an NDI may be defined for one TB.
Although it is be assumed herein that independent NDIs are defined
for individual TBs, embodiments of the present invention may be
applied to the case where only one NDI is defined for one TB.
FIG. 7 illustrates a general operation of a Node B in a wireless
communication system supporting UL MIMO, in which it is assumed
that two TBs are transmitted. A grant in a PUSCH carrying only one
TB has been described with reference to FIG. 3.
Referring to FIG. 7, a Node B performs UL scheduling in step 701,
to determine a UE the Node B will grant to transmit a PUSCH and
also to determine a resource the UE will use for the PUSCH
transmission. In step 703, the Node B transmits a PDCCH to the
scheduled UE to provide grant information for initial transmission
of a PUSCH. Since UL MIMO transmission is considered herein, the
grant information indicates that a rank is greater than or equal to
2 (rank>1). Specifically, it is assumed that transmission of two
TBs is granted. Since the grant is for initial transmission of a
PUSCH, values of NDIs corresponding to the TBs are toggled. In step
705, the Node B demodulates and decodes the PUSCH that has been
received four subframes after the time the PDCCH was transmitted in
step 703. In step 707, the Node B determines if the decoding of the
PUSCH is successful. Because two TBs were transmitted, there are
four possible cases where success/failure in the decoding may be
determined. Case 1: Both of TB1 and TB2 have been successfully
decoded. In this case, the Node B transmits (ACK, ACK) for TB1 and
TB2 using a PHICH in step 709, and then returns to step 701. Case
2: TB1 has been successfully decoded but TB2 has been failed to be
successfully decoded. In this case, the Node B transmits (ACK,
NACK) for TB1 and TB2 using a PHICH in step 711, and transmits a
PDCCH for granting PUSCH retransmission, to the UE in step 715.
Case 3: TB2 has been successfully decoded but TB1 has been failed
to be successfully decoded. In this case, the Node B transmits
(NACK, ACK) for TB1 and TB2 using a PHICH in step 713, and
transmits a PDCCH for granting PUSCH retransmission, to the UE in
step 715. Case 4: Both of TB1 and TB2 have failed to be
successfully decoded. In this case, the Node B transmits (NACK,
NACK) for TB1 and TB2 using a PHICH in step 717, and determines in
step 719 whether to change the transmission property for the
initial transmission that the Node B informed the UE in step 703,
during retransmission of the PUSCH. When determining to change the
transmission property, the Node B returns to step 705 after
performing step 715. When determining not to change the
transmission property for the initial transmission, the Node B
returns to step 705 to receive and decode the PUSCH, supposing that
the UE retransmits the PUSCH, maintaining the transmission property
for the initial transmission. Step 715 corresponds to a process in
which the Node B transmits a PDCCH to inform the UE of the
transmission property to be used for transmission of the PUSCH.
After returning to step 705, the Node B receives and decodes the
PUSCH received from the UE, considering that the UE retransmits the
PUSCH according to the transmission property indicated in step
715.
In the foregoing description of FIG. 7, in Case 1 or Case 4, the
Node B may instruct retransmission of the PUSCH simply by
transmitting a PHICH without transmitting a PDCCH. However, in Case
2 or 3 where only one of the two TBs has been successfully decoded,
the Node B should transmit a PDCCH indicating the transmission
property in order to instruct retransmission of the PUSCH. The
reason for transmitting this PDCCH is set forth below.
As summarized in Table 1, the number of TBs to be transmitted is
changed according to the rank value. For example, while
transmitting two TBs in initial transmission, if a UE does not need
to transmit them as one of the two TBs has been successfully
decoded, the UE transmits one TB in retransmission. If the number
of TBs to be transmitted is reduced, a rank value in retransmission
is smaller than that in the initial transmission. However, as shown
in Tables 2 and 3, different precoders are defined for different
ranks. So, the precoders used in initial transmission may not be
used in retransmission. For this reason, if the number of TBs
transmitted in retransmission is reduced, a PDCCH should be
transmitted to inform the changed precoder to be used by the UE,
i.e., the changed transmission property.
FIG. 8 illustrates a general operation of a UE in a wireless
communication system supporting UL MIMO, in which the UE operation
corresponds to the Node B operation in FIG. 7.
Referring to FIG. 8, a UE attempts to receive and decode a PDCCH
for a UL grant in step 801, and determines in step 803 whether it
has succeeded in decoding. If the UE has successfully decoded the
PDCCH for a UL grant, the UE determines in step 805 whether an NDI
has been toggled. Since UL MIMO transmission considered, it will be
assumed herein that an initial grant indicates information about
two TBs. Assuming that different NDIs are defined for different
TBs, if both of the two NDIs have not been toggled from their
previous values, corresponding to a UL grant indicating mere
retransmission, then the UE retransmits the PUSCH by reflecting the
new transmission property including a PMI in step 807. However, if
any one of the two NDIs has been toggled, the UE proceeds to step
809, in which the TB corresponding to the toggled NDI should be
newly transmitted, and the TB corresponding to the non-toggled NDI
should be retransmitted. No matter whether each TB is subjected to
retransmission or initial transmission, the transmission property
including a PMI should follow the value indicated in the PDCCH.
Even though only one NDI is defined regardless of the number of
TBs, if the NDI has been toggled from its previous value, the UE
performs step 809 to transmit a new TB. Otherwise, the UE performs
step 807, for retransmission. In summary, in terms of precoding,
upon receiving a PDCCH, a UE is allowed to transmit a PUSCH
precoded by reflecting a PMI indicated in the PDCCH, regardless of
whether the TB is for retransmission.
If the UE has failed to receive and decode a PDCCH in step 803, the
UE attempts to receive and decode a PHICH in step 811. If ACK/NACK
information for each TB is present in the PHICH, the UE may respond
differently to the following three cases.
Case 1: ACKs are received for both of the two TBs. In this case,
the UE does not need to transmit a PUSCH in step 815.
Case 2: An ACK is received for one TB, and a NACK is received for
the other TB. In this case, the UE should inevitably change the
rank value, because of a reduction in the number of TBs to be
retransmitted. However, since a Node B has not separately provided
a PMI (since the UE has failed to receive a PDCCH), the UE may not
determine a precoding scheme to be used for PUSCH transmission.
Therefore, the UE may not define its PUSCH transmission operation
in step 819.
Case 3: NACKs are received for both of the two TBs, and the two TBs
need to be retransmitted. In this case, because the UE has no
change in rank and the Node B has not separately provided a PMI,
the UE is allowed to retransmit the PUSCH by reflecting the
transmission property including the PMI indicated in the last
received UL grant, in step 817. In this case, however, the UE
changes an RV in retransmission according to the rule of
Incremental Redundancy (IR) synchronous HARQ. In a PUSCH
retransmitted by the PHICH, an RV automatically increases without a
separate instruction. As is well known in the art, an HARQ
retransmission scheme is classified into Chase Combining (CC) and
Incremental Redundancy (IR) schemes. CC is a scheme of transmitting
the same signals both in retransmission and initial transmission so
that a receiver may combine the signals in a symbol level. IR is a
scheme of transmitting signals with different RVs in retransmission
and initial transmission so that a receiver may combine the signals
in a decoding process. IR is popularly used as a HARQ
retransmission scheme despite its high reception complexity
compared to CC, because it may additionally obtain a decoding gain.
In synchronous HARQ, an RV is implicitly determined because a
separate PDCCH for changing an RV is not transmitted in
retransmission. For example, in the LTE system, a total of four RVs
are defined (RV=0, 1, 2, 3), and if synchronous HARQ is applied,
RVs are applied in order of {0,2,3,1} according to their
transmission order.
In this general wireless communication system supporting UL MIMO,
if some of multiple received TBs have been failed to be
successfully decoded (e.g., only one of two TBs has been
successfully decoded), a Node B should transmit a PDCCH indicating
the transmission property in order to instruct a UE to retransmit
the PUSCH. Otherwise, the UE may not determine the precoding scheme
to be used for PUSCH transmission.
However, as described above, the transmission of a PDCCH may
increase resource consumption for control information. Therefore,
to reduce the transmission load regarding control information for
retransmission in a UL, embodiments of the present invention
provide a method for controlling UL HARQ only with a PHICH and
determining a precoding scheme to be used in retransmission by a UE
without transmission of a PDCCH in the LTE system supporting UL
MIMO.
Structures of a UE's transmission device and a Node B's reception
device, to which the method of determining a precoding scheme to be
used in retransmission according to embodiments of the present
invention are applied, are as shown in FIGS. 5 and 6, respectively.
The UE's controller 241 and the Node B's controller 313 in FIGS. 5
and 6 determine a precoding scheme to be used in retransmission
according to the procedures of FIGS. 10 and 9, respectively.
FIG. 9 illustrates a procedure for determining a precoding scheme
in a Node B supporting UL MIMO, according to an embodiment of the
present invention. The procedure of FIG. 9 partially overlaps the
conventional procedure of FIG. 7. Therefore, the following
description will be focused on the difference between the procedure
of FIG. 9 and the conventional procedure of FIG. 7.
Referring FIG. 9, steps 901 to 907 in which a Node B performs UL
scheduling to grant initial transmission of a PUSCH,
receives/decodes the PUSCH transmitted by a UE, and determines the
success/failure in decoding, are identical in operation to steps
701 to 707 in FIG. 7.
In step 907, the Node B determines if the decoding of the PUSCH is
successful. Assuming that a UE has transmitted two TBs, there are
four possible cases where the success/failure in decoding may be
determined. In Case 1 where both of the two TBs have been
successfully decoded, and Case 4 where both of the two TBs have
been failed to be successfully decoded, the Node B proceeds to
steps 909 and 915, respectively, and operates as in FIG. 7.
However, in Cases 2 and 3 where one TB has been successfully
decoded but the other TB has failed to be successfully decoded, the
Node B proceeds to steps 911 and 913, respectively, and then
performs step 917. In the conventional procedure of FIG. 7, step
917 (or 719) is performed only in Case 4 where both of the two TBs
have failed to be successfully decoded. However, in the present
invention, step 917 is performed even in Cases 2 and 3 where one of
the two TBs has failed to be successfully decoded. In step 917, the
Node B determines whether to change the transmission property for
the initial transmission that the Node B informed the UE in step
903, in retransmission of the PUSCH. When determining to change the
transmission property, the Node B performs step 919. Otherwise, the
Node B performs step 921.
Embodiments of the present invention provide methods for implicitly
determining which precoding it will use in the retransmission
situation. In step 921, the Node B determines whether it will
maintain the implicitly defined precoding, or indicate another
precoding. When determining to maintain the implicitly defined
precoding, the Node B returns to step 905 to receive and decode a
PUSCH to which the implicitly defined precoding is applied, because
there is no reason to transmit a PDCCH.
However, when determining to indicate a specific precoding scheme,
instead of using the implicitly determined precoding, the Node B
performs step 919 to specifically indicate the transmission
property needed for PUSCH transmission through a PDCCH. The method
of using the implicitly determined precoding is a method of
determining precoding to be used during retransmission without
transmission of the PDCCH, and will be described in greater detail
in the following embodiments.
If the Node B indicates the resource-saving effects that when using
the implicitly defined precoding, the Node B is not required to
transmit a PDCCH, and also indicates specific precoding in step
921, the Node B determines whether to transmit a PDCCH by comparing
the resource-saving effects with the advantages of AMC by which
precoding most appropriate for the UL channel state can be applied.
In this way, the present invention may enable the Node B to freely
operate resources.
FIG. 10 illustrates a procedure for determining a precoding scheme
in a UE supporting UL MIMO, according to an embodiment of the
present invention.
The procedure of FIG. 10 partially overlaps the conventional
procedure of FIG. 8. Therefore, the following description will be
focused on the difference between the procedure of FIG. 10 and the
conventional procedure of FIG. 8.
In the conventional procedure of FIG. 8, if an ACK is received for
one TB and a NACK is received for the other TB from a PHICH, an
operation of the UE may not be defined (step 819 in FIG. 8). In
this case, however, embodiments of the present invention perform
step 1019 in FIG. 10. Steps 1001 to 1017 in FIG. 10 are identical
in operation to steps 801 to 817 in FIG. 8. In steps 1001 to 1017,
the UE attempts to receive and decode a PDCCH for a UL grant,
determines if NDIs have been toggled, if the decoding of the PDCCH
is successful, transmits a PUSCH precoded by reflecting a PMI
indicated in the PDCCH, receives a PHICH if the decoding of the
PDCCH is failed (or no PDCCH is transmitted), transmits a PUSCH if
ACK/NACK information about TBs, included in the PHICH, indicates
that ACKs are received for both of the two TBs, and retransmits the
PUSCH by reflecting the transmission property indicated in the last
received UL grant, if NACKs are received for both of the two
TBs.
In step 1019, in retransmitting the PUSCH, the UE maintains all the
transmission properties except for the precoding and the RV at
values of the last received UL grant, determines the RV according
to the conventional rule, and uses the implicitly determined
precoding scheme. Regarding the method of implicitly determining
the precoding to be used in retransmission according to the present
invention, the related agreement is made between a Node B and a UE
in advance. Therefore, if the UE receives an ACK for one TB and a
NACK for the other TB, the UE may perform PUSCH retransmission by
UL MIMO by merely receiving a PHICH in step 1019.
The method of implicitly determining the precoding to be used in
retransmission will be described in detail below in the following
embodiments. In accordance with the precoding determining method
proposed in the following embodiments, the Node B performs step 921
in FIG. 9, and the UE performs step 1019 in FIG. 10.
FIG. 11 illustrates a precoding determining method performed in a
UE during retransmission, according to a first embodiment of the
present invention.
Referring to FIG. 11, a UE receives and decodes a PHICH in step
1101, and determines in step 1103 whether ACK/NACK information
carried by the PHICH is an ACK. The determination results are
classified into three different cases. In Case 1, where ACKs are
received for both of two TBs, the UE stops retransmission of a
PUSCH and does not require precoding information in step 1105. In
Case 2, where an ACK is received for one TB but a NACK is received
for the other TB, the UE specifically determines for which TB it
has received an ACK in step 1109. In Case 3, where NACKs are
received for both of the two TBs, the UE determines to reuse the
precoding indicated in an initial grant in retransmission in step
1107.
In step 1109, if an ACK is received for one TB but a NACK is
received for the other TB, the UE determines for which TB it has
received an ACK. In Case 1 where (ACK, NACK) is identified, it is
assumed that an ACK is identified for TB1 and a NACK is identified
for TB2. In this case, the UE performs step 1111 and its succeeding
steps. In Case 2 where (NACK, ACK) is identified, it is assumed
that a NACK is identified for TB1 and an ACK is identified for TB2.
In this case, the UE performs step 1117 and its succeeding
steps.
Step 1111 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH. An initial rank is always
greater than or equal to 2, because it is assumed herein that in
initial transmission, two TBs are transmitted by UL MIMO. In Case
1-1 where it is determined in step 1111 that an initial rank is 2,
the UE performs step 1113, and in Case 1-2 where an initial rank
exceeds 2, the UE performs step 1115. In step 1113, the UE
implicitly determines the last column vector in the precoding
matrix used in initial transmission as a precoding matrix for
retransmission (or a retransmission precoding matrix) without
receiving a PDCCH. As shown in Table 1, if two TBs are transmitted,
TB1 uses the first one or two column vectors in the precoding
matrix, while TB2 uses the last one or two column vectors in the
precoding matrix. An operation following step 1111 is performed in
the (ACK, NACK) situation, and since TB2 should be retransmitted,
the last column vector(s) are used for retransmission. Step 1113 is
performed in Case 1-1 where an initial rank is 2, and if a rank is
2, each TB occupies one layer in initial transmission. Therefore,
because TB2 should occupy one layer even in retransmission, only
the last column vector is taken and determined as a retransmission
precoding matrix. In step 1115, the last two column vectors in the
precoding matrix used in initial transmission are determined as a
retransmission precoding matrix. Step 1115 is performed when an
initial rank is 3 or 4, and in this case, TB2 occupies two layers
in initial transmission as in Table 1. Therefore, the last two
column vectors are taken and determined as a precoding matrix so
that TB2 may occupy two layers even in retransmission.
In Case 2 where it is determined in step 1109 that ACK/NACK
information carried by the PHICH is identified as (NACK, ACK), the
UE determines which rank was used in initial transmission of a
PUSCH in step 1117. An initial rank is always greater than or equal
to 2, because it is assumed herein that in initial transmission,
two TBs are transmitted by UL MIMO. In Case 2-1 where it is
determined in step 1117 that an initial rank is less than 4, the UE
performs step 1119, and in Case 2-2 where an initial rank is 4, the
UE performs step 1121. Step 1119 is for determining the first
column vector in the precoding matrix used in initial transmission
as a retransmission precoding matrix. An operation following step
1119 is performed in the (NACK, ACK) situation, and since TB1
should be retransmitted, the first column vector(s) in the
precoding matrix are used for retransmission. Step 1119 is
performed when an initial rank is 2 or 3, and if a rank is 2 or 3,
TB1 occupies one layer in initial transmission. Therefore, since
TB1 should occupy one layer even in retransmission, only the first
column vector is taken and determined as a retransmission precoding
matrix. Step 1121 is for determining the first two column vectors
in the precoding matrix used in initial transmission as a
retransmission precoding matrix. Step 1121 is performed when an
initial rank is 4, and in this case, TB1 occupies two layers in
initial transmission as in Table 1. Therefore, the first two column
vectors are taken and determined as a retransmission precoding
matrix so that TB1 may occupy two layers even in
retransmission.
In summary, the precoding matrix used in initial transmission is
reused intact in retransmission, but no signal is transmitted over
the layer for transmitting the successfully received TB. Therefore,
in the first embodiment of the present invention, the method of
determining a retransmission precoding matrix may be called a
method of blanking a layer for the successfully received TB.
A specific example of the first embodiment of the present invention
is described below. It is assumed that in initial transmission, a
matrix P.sub.0 shown in Equation (2) below is used as a precoding
matrix.
.function. ##EQU00062##
In initial transmission, TB1 is precoded using a matrix P.sub.1 in
Equation (3) below, while TB2 is precoded using a matrix P.sub.2 in
Equation (3).
.function..function. ##EQU00063##
Upon receiving (NACK, ACK) from a PHICH, the UE uses the matrix
P.sub.1 as a retransmission precoding matrix since it should
retransmit TB1. Upon receiving (ACK, NACK), the UE uses the matrix
P.sub.2 as a retransmission precoding matrix since it should
retransmit TB2.
After implicitly determining the precoding matrix to be used in
retransmission, the UE boosts its transmission power by XdB for
PUSCH transmission in step 1123. Specifically, the UE performs
implicit power boosting. A level by which the transmission power is
boosted is variable depending on the precoding matrix to be used.
According to the first embodiment of the present invention, in
retransmission, not all transmission antennas of the UE can be used
for retransmission. Therefore, the number of antennas used may
decrease in retransmission. Actually, the UE cannot use more
transmission antennas even though it may utilize more transmission
power if using more transmission antennas. To solve this problem,
during retransmission, transmission power may be implicitly boosted
as much as the ratio of the number of antennas used in
retransmission to the number of antennas used in initial
transmission. For example, if the number of antennas used in
initial transmission is 4 and the number of antennas used in
retransmission is 2, the ratio is 4/2=2, making it possible to
boost the transmission power by 3 dB(X=3). Although the method of
performing implicit power boosting during PUSCH transmission has
been described as an example of the embodiment of the present
invention, another method of not boosting transmission power of a
PUSCH to reduce interference to other users even in the conditions
in which the transmission power of the PUSCH can be boosted, may be
defined as another example.
FIG. 12 illustrates a precoding determining method performed in a
Node B during retransmission according to the first embodiment of
the present invention, in which the procedure of FIG. 12 is assumed
to be performed in step 921 in FIG. 9. It is to be noted that a
method of determining (assuming) a precoding scheme used by a UE
during retransmission by a Node B in steps 1205 to 1221 in FIG. 12
is identical to the method of determining (defining) a precoding
scheme by a UE in steps 1105 to 1121 in FIG. 11.
Referring to FIG. 12, a Node B receives and decodes a PUSCH
transmitted by a UE in step 1201, and determines in step 1203
whether the decoding of the PUSCH is successful by determining
whether ACK/NACK information transmitted to the UE over a PHICH is
an ACK or a NACK. The determination results are classified into
three different cases. In Case 1, where (ACK, ACK) are transmitted
in step 909 in FIG. 9 as both of two TBs have been successfully
decoded, the Node B assumes that the UE will stop retransmission of
the PUSCH, and does not need precoding information, in step 1205.
In Case 2, where an ACK is transmitted for one TB but a NACK is
transmitted for the other TB as in steps 911 and 913 in FIG. 9, the
Node B specifically determines for which TB it has transmitted an
ACK in step 1209. In Case 3, where (NACK, NACK) are transmitted for
both TBs in step 915 in FIG. 9 as both of the two TBs have been
failed to be successfully decoded, the Node B assumes in step 1207
that the UE will reuse the precoding indicated in an initial grant
in retransmission.
In step 1209, the Node B determines for which TB it has transmitted
an ACK, when it transmitted an ACK for one TB and a NACK for the
other TB. In Case 1, where (ACK, NACK) is identified, it is assumed
that an ACK is identified for TB1 and a NACK is identified for TB2.
In this case, the Node B performs step 1211 and its succeeding
steps. In Case 2, where (NACK, ACK) is identified, it is assumed
that a NACK is identified for TB1 and an ACK is identified for TB2.
In this case, the Node B performs 1217 and its succeeding
steps.
Step 1211 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 1. An initial rank
is always greater than or equal to 2, because it is assumed herein
that in initial transmission, two TBs are transmitted by UL MIMO.
In Case 1-1 where it is determined in step 1211 that an initial
rank is 2, the Node B performs step 1213, and in Case 1-2 where an
initial rank exceeds 2, the Node B performs step 1215. In step
1213, the Node B implicitly determines (assumes) the last column
vector in the precoding matrix used in initial transmission as a
retransmission precoding matrix used by the UE. In step 1215, the
Node B implicitly determines the last two column vectors in the
precoding matrix used in initial transmission as a retransmission
precoding matrix used by the UE.
On the other hand, in Case 2, the Node B determines which rank was
used in initial transmission of a PUSCH, in step 1217. An initial
rank is always greater than or equal to 2, because it is assumed
herein that in initial transmission, two TBs are transmitted by UL
MIMO. In Case 2-1 where it is determined in step 1217 that an
initial rank is less than 4, the Node B performs step 1219, and in
Case 2-2 where an initial rank is 4, the Node B performs step 1221.
In step 1219, the Node B implicitly determines the first column
vector in the precoding matrix used in initial transmission as a
retransmission precoding matrix. In step 1221, the Node B
implicitly determines the first two column vectors in the precoding
matrix used in initial transmission as a retransmission precoding
matrix.
The precoding matrixes for retransmission are determined as in
steps 1213, 1215, 1219, and 1221 for reasons described in detail
with reference to FIG. 11.
As described in connection with FIGS. 11 and 12, as a UE and a Node
B implicitly determine in the same way the precoding scheme to be
used during retransmission if ACK/NACK information for, for
example, two TBs is identified as (ACK, NACK) or (NACK, ACK), they
are not required to transmit/receive a PDCCH indicating a precoding
scheme during retransmission, thereby making it possible to reduce
the transmission load caused by frequent transmission and reception
of a PDCCH.
The basic concept of a second embodiment of the present invention
is that a precoding matrix to be used in retransmission is
determined in advance. To this end, in the second embodiment of the
present invention, the concept of a default codebook is newly
defined and one of the precoding matrixes defined in the default
codebook is selected and used for retransmission. One or more
precoding matrixes may be defined in the default codebook. If only
one precoding matrix is defined in the default codebook, only the
precoding matrix can be used in retransmission. Otherwise, if
multiple precoding matrixes are defined in the default codebook, a
precoding matrix is selected and used from among the multiple
precoding matrixes according to a predetermined rule in
retransmission. Regarding a rule of selecting a precoding matrix in
retransmission, the following methods are available.
i. Selecting a precoding matrix according to an RV value: In the
LTE system, since a total of four RVs are defined, a default
codebook using this rule may include four precoding matrixes. In
the LTE system, UL HARQ is set to transmit a different RV in every
retransmission, thereby obtaining the effect of applying a
different precoding matrix in every retransmission. This rule is to
obtain the possible maximum spatial diversity gain in
retransmission considering that AMC is not appropriately operating,
if a TB has not been successfully decoded in initial
transmission.
ii. Selecting a precoding matrix according to a system frame number
or a subframe number: In the LTE system, system frame numbers and
subframe numbers are defined to number resources in the time
domain. In the LTE system, a system frame, which is a 10-ms
resource, includes 10 subframes. The subframe is a 1-ms resource,
and a subframe number is initialized in every system frame. For
example, assume that Q precoding matrixes are defined in a default
codebook. If a system frame number of the time resource at which
retransmission occurs is represented by n.sub.SFN, and a subframe
number is represented by n, then k is calculated in accordance with
Equation (4) below, and a k-th matrix in the default codebook is
used as a precoding matrix. Here, `mod(A, B)` represents a
remainder obtained by dividing A by B. If only the system frame
number is used as an input factor for determining a precoding
matrix, `k=mod(n.sub.SFN, Q)` is used to determine a precoding
matrix. If only the subframe number is used as an input factor for
determining a precoding matrix, `k=(n, Q)` is used to determine a
precoding matrix. This rule is also for obtaining a spatial
diversity gain in retransmission. k=mod(10.times.n.sub.SFN+n,Q)
(4)
FIG. 13 illustrates a precoding determining method performed in a
UE during retransmission, according to the second embodiment of the
present invention, in which steps 1301 to 1317 are identical in
operation to steps 1101 to 1117 in FIG. 11.
Referring to FIG. 13, in Case 1-1, where it is determined in step
1311 that a UE has received (ACK, NACK) from a PHICH and an initial
rank was 2, the UE determines to use a precoding matrix in an
A-type default rank-1 codebook in step 1313. In Case 1-2, where it
is determined in step 1311 that the UE has received (ACK, NACK)
from a PHICH and an initial rank was exceeding 2, the UE determines
to use a precoding matrix in an A-type default rank-2 codebook in
step 1315. An A-type default codebook is for retransmitting a
second TB. A method of selecting a precoding matrix in a default
codebook follows any one of the rule (i) of selecting a precoding
matrix according to the RV value, and the rule (ii) of selecting a
precoding matrix according to the system frame number or the
subframe number.
In Case 2-1 where it is determined in step 1317 that the UE has
received (NACK, ACK) from a PHICH and an initial rank is less than
4, the UE determines to use a precoding matrix in a B-type default
rank-1 codebook in step 1319. In Case 2-2 where it is determined in
step 1317 that the UE has received (NACK, ACK) from a PHICH and an
initial rank is 4, the UE determines to use a precoding matrix in a
B-type default rank-2 codebook in step 1321. A B-type default
codebook is used for retransmission of a first TB, and a method of
selecting a precoding matrix in a default codebook follows any one
of the rules (i) and (ii).
While the A-type default codebook and the B-type default codebook
are separately defined in the foregoing description, the A-type
default codebook and the B-type default codebook may be designed to
be identical to each other, if the default codebooks are defined
regardless of which TB is retransmitted by the default codebooks.
Instead, a rule of specifically determining which precoding matrix
in a default codebook is to be used in retransmission may be
different for every TB.
In summary, using a function of an RV or a time resource number, a
precoding matrix is selected in a default codebook that is defined
according to which rank transmission is performed and which TB is
transmitted in retransmission regardless of the precoding matrix
used in initial transmission. Therefore, this method may be called
a default codebook method.
FIG. 14 illustrates a precoding determining method performed in a
Node B during retransmission, according to the second embodiment of
the present invention.
The procedure of FIG. 14 is assumed to be performed in step 921 of
FIG. 9. It is to be noted that a method of determining (assuming) a
precoding scheme used by a UE during retransmission by a Node B in
steps 1405 to 1421 in FIG. 14 is identical to the method of
determining (defining) a precoding scheme by a UE in steps 1305 to
1321 in FIG. 13. Since steps 1401 to 1417 in FIG. 14 are identical
in operation to steps 1201 to 1217 in FIG. 12, a detailed
description thereof will be omitted.
Referring to FIG. 14, in Case 1-1, where it is determined in step
1411 that decoding results on two TBs are (ACK, NACK) and an
initial rank was 2, the Node B determines (assumes) in step 1413
that a UE uses a precoding matrix in an A-type default rank-1
codebook. In Case 1-2, where it is determined in step 1411 that
decoding results on two TBs are (ACK, NACK) and an initial rank
exceeds 2, the Node B determines (assumes) in step 1415 that a UE
uses a precoding matrix in an A-type default rank-2 codebook.
On the other hand, in Case 2-1, where it is determined in step 1417
that decoding results on two TBs are (NACK, ACK) and an initial
rank is less than 4, the Node B determines (assumes) in step 1419
that a UE uses a precoding matrix in a B-type default rank-1
codebook. In Case 2-2, where it is determined in step 1417 that
decoding results on two TBs are (NACK, ACK) and an initial rank is
4, the Node B determines (assumes) in step 1421 that a UE uses a
precoding matrix in a B-type default rank-2 codebook.
An A-type default codebook is used for retransmission of a second
TB, while a B-type default codebook is used for retransmission of a
first TB. A method of selecting a precoding matrix in the A-type
default codebook and the B-type default codebook by a Node B
follows any one of the rule (i) of selecting a precoding matrix
according to the RV value and the rule (ii) of selecting a
precoding matrix according to the system frame number or the
subframe number.
The basic concept of a third embodiment is that a precoding matrix
to be used in retransmission is selected in a codebook having a
functional relationship with a precoding matrix indicated in an
initial grant of a PUSCH. For the functional relationship, various
known functional formulae applicable to matrixes may be used. To
this end, in this embodiment of the present invention, the
`mother-child pair` concept is newly defined in precoding matrixes.
A precoding matrix indicated in an initial grant of a PUSCH serves
as a mother precoding matrix, and one of the precoding matrixes in
a child codebook defined by a functional relationship determined
from a function of the mother precoding matrix is selected and used
for retransmission. One or more precoding matrixes may be defined
in the default codebook. If only one precoding matrix is defined in
the child codebook, only the precoding matrix can be used in
retransmission. Otherwise, if multiple precoding matrixes are
defined in the child codebook, a precoding matrix is selected and
used from among the child codebook according to a predetermined
rule in retransmission. As to a rule of selecting a precoding
matrix in retransmission, the following rules mentioned in the
second embodiment may be used.
i. Selecting a precoding matrix according to an RV value; and
ii. Selecting a precoding matrix according to a system frame number
or a subframe number
The technical basis of using the child codebook is as follows. If a
change in the spatial signature of a channel was not significant
between initial transmission and retransmission, using a child
precoding matrix similar to a mother precoding matrix will help
increase a MIMO gain. Therefore, a functional relationship for
enabling decision on a child precoding matrix is given according to
what mother precoding matrix was used. If a MIMO gain was not high
despite a definition of one child precoding matrix, multiple child
precoding matrixes may be defined as a child codebook and one of
the child precoding matrixes may be selected and used in
retransmission according to the above rules.
FIG. 15 illustrates a precoding determining method performed in a
UE during retransmission, according to the third embodiment of the
present invention. Steps 1501 to 1517 in FIG. 15 are identical in
operation to steps 1101 to 1117 in FIG. 11, so a detailed
description thereof will be omitted.
Referring to FIG. 15, in Case 1-1, where it is determined in step
1511 that a UE has received (ACK, NACK) from a PHICH and an initial
rank was 2, the UE determines to use a precoding matrix in an
A-type rank-1 child codebook in step 1513. As described above, a
child codebook is defined from a precoding matrix used in initial
transmission using a predetermined functional relationship. In Case
1-2, where it is determined in step 1511 that the UE has received
(ACK, NACK) from a PHICH and an initial rank exceeds 2, the UE
determines to use a precoding matrix in an A-type rank-2 child
codebook in step 1515. An A-type child codebook is to be used for
retransmission of a second TB. A method of selecting a precoding
matrix in a child codebook follows any one of the rules (i) and
(ii).
In Case 2-1, where it is determined in step 1517 that the UE has
received (NACK, ACK) from a PHICH and an initial rank is less than
4, the UE determines to use a precoding matrix in a B-type rank-1
child codebook in step 1519. In Case 2-2, where it is determined in
step 1517 that the UE has received (NACK, ACK) from a PHICH and an
initial rank is 4, the UE determines to use a precoding matrix in a
B-type rank-2 child codebook in step 1521. A B-type child codebook
is used for retransmission of a first TB. A method of selecting a
precoding matrix in a child codebook follows any one of the rules
(i) and (ii).
While the A-type child codebook and the B-type child codebook are
separately defined in the foregoing description, the A-type child
codebook and the B-type child codebook may be designed to be
identical to each other, if the child codebooks are defined
regardless of which TB is retransmitted by the child codebooks.
Instead, a rule of specifically determining which precoding matrix
in a child codebook is to be used in retransmission may be
different for every TB.
In summary, using a function of an RV or a time resource number
(e.g., a system frame number and a subframe number), a precoding
matrix is selected in a child codebook that is determined according
to a precoding matrix used in initial transmission, and which rank
transmission is performed and which TB is transmitted in
retransmission. Therefore, this method may be called a mother-child
pair method.
FIG. 16 illustrates a precoding determining method performed in a
Node B during retransmission, according to the third embodiment of
the present invention.
The procedure of FIG. 16 is assumed to be performed in step 921 of
FIG. 9. It is to be noted that a method of determining (assuming) a
precoding scheme used by a UE during retransmission by a Node B in
steps 1605 to 1621 in FIG. 16 is identical to the method of
determining (defining) a precoding scheme by a UE in steps 1505 to
1521 in FIG. 15. Since steps 1601 to 1617 in FIG. 16 are identical
in operation to steps 1201 to 1217 in FIG. 12, a detailed
description thereof will be omitted.
Referring to FIG. 16, in Case 1-1, where it is determined in step
1611 that decoding results on two TBs are (ACK, NACK) and an
initial rank was 2, a Node B determines (assumes) in step 1613 that
a UE uses a precoding matrix in an A-type rank-1 child codebook. As
described above, a child codebook is defined from a precoding
matrix used in initial transmission using a predetermined
functional relationship. In Case 1-2, where it is determined in
step 1611 that decoding results on two TBs are (ACK, NACK) and an
initial rank exceeds 2, the Node B determines (assumes) in step
1615 that a UE uses a precoding matrix in an A-type rank-2 child
codebook.
In Case 2-1, where it is determined in step 1617 that decoding
results on two TBs are (NACK, ACK) and an initial rank is less than
4, the Node B determines (assumes) in step 1619 that a UE uses a
precoding matrix in a B-type rank-1 child codebook. In Case 2-2,
where it is determined in step 1617 that decoding results on two
TBs are (NACK, ACK) and an initial rank is 4, the Node B determines
(assumes) in step 1621 that a UE uses a precoding matrix in a
B-type rank-2 child codebook.
An A-type child codebook is used for retransmission of a second TB,
while a B-type child codebook is used for retransmission of a first
TB. A method of selecting a precoding matrix in the A-type child
codebook and the B-type child codebook by a Node B follows any one
of the rule (i) of selecting a precoding matrix according to the RV
value and the rule (ii) of selecting a precoding matrix according
to the system frame number or the subframe number.
The basic concept of a fourth embodiment of the present invention
is that a precoding matrix to be used in retransmission is informed
(indicated) together in an initial grant of a PUSCH. In the
conventional method, a precoding matrix to be used in initial
transmission is indicated by a PMI in an initial grant. However, in
the fourth embodiment of the present invention, in addition to the
precoding matrix to be used in initial transmission, a precoding
matrix to be used when retransmission should be performed
responding only to a PHICH, or its candidate group, may be
additionally indicated by a PMI in an initial grant. A candidate
precoding matrix group for retransmission, indicated by a PMI in an
initial grant, will be referred to as a codebook for retransmission
(or a retransmission codebook). One or more precoding matrixes may
be defined in a retransmission codebook. If only one precoding
matrix is defined in a retransmission codebook, only the precoding
matrix can be used in retransmission. Otherwise, if multiple
precoding matrixes are defined in a retransmission codebook, a
precoding matrix is selected and used according to a specific rule
in retransmission. As to a rule of selecting a precoding matrix in
retransmission, the following rules mentioned in the second
embodiment may be used.
i. Selecting a precoding matrix according to an RV value; and
ii. Selecting a precoding matrix according to a system frame number
or a subframe number
The technical basis of using the retransmission codebook is similar
to that of the third embodiment of the present invention.
Specifically, if a change in the spatial signature of a channel was
not significant between initial transmission and retransmission,
using a precoding matrix similar to the precoding matrix used in
initial transmission, for retransmission, will help increase a MIMO
gain. However, a Node B may best know which precoding matrix it
should use to help increase a MIMO gain. Therefore, a
retransmission codebook is indicated together in an initial
grant.
FIG. 17 illustrates a precoding determining method performed in a
UE during retransmission, according to the fourth embodiment of the
present invention.
Steps 1701 to 1717 in FIG. 17 are identical in operation to steps
1101 to 1117 in FIG. 11, so a detailed description thereof will be
omitted.
Referring to FIG. 17, in Case 1-1, where it is determined in step
1711 that a UE has received (ACK, NACK) from a PHICH and an initial
rank was 2, the UE determines to use a precoding matrix in an
A-type rank-1 retransmission codebook in step 1713. As described
above, the A-type rank-1 retransmission codebook is assumed to be
indicated in an initial grant. In Case 1-2, where it is determined
in step 1711 that the UE has received (ACK, NACK) from a PHICH and
an initial rank exceeds 2, the UE determines to use a precoding
matrix in an A-type rank-2 retransmission codebook in step 1715.
Likewise, as described above, the A-type rank-2 retransmission
codebook is assumed to be indicated in an initial grant. An A-type
retransmission codebook is used for retransmission of a second TB.
In a method of selecting a precoding matrix in a retransmission
codebook, any one of the rules described in the second embodiment
of the present invention may be used.
In Case 2-1, where it is determined in step 1717 that the UE has
received (NACK, ACK) from a PHICH and an initial rank is less than
4, the UE determines to use a precoding matrix in a B-type rank-1
retransmission codebook in step 1719. As described above, the
B-type rank-1 retransmission codebook is assumed to be indicated in
an initial grant. In Case 2-2, where it is determined in step 1717
that the UE has received (NACK, ACK) from a PHICH and an initial
rank is 4, the UE determines to use a precoding matrix in a B-type
rank-2 retransmission codebook in step 1721. Likewise, as described
above, the B-type rank-2 retransmission codebook is assumed to be
indicated in an initial grant. A B-type retransmission codebook is
used for retransmission of a first TB. In a method of selecting a
precoding matrix in a retransmission codebook, any one of the rules
described in the second embodiment of the present invention may be
used.
While the A-type retransmission codebook and the B-type
retransmission codebook are separately defined in the foregoing
description, the A-type retransmission codebook and the B-type
retransmission codebook may be designed to be identical to each
other, if the retransmission codebooks are defined regardless of
which TB is retransmitted by the retransmission codebooks. Instead,
a rule of specifically determining which precoding matrix in a
retransmission codebook is to be used in retransmission may be
different for every TB.
In summary, a candidate precoding matrix group for retransmission
is determined by a PMI value indicated in an initial grant, and
using a function of an RV or a time resource number (e.g., a system
frame number and a subframe number), a precoding matrix is selected
in a retransmission codebook that is determined according to which
rank transmission is performed and which TB is transmitted in
retransmission. Therefore, this method may be called a
PDCCH-indicating method.
TABLE-US-00004 TABLE 4 Examples of Retransmission Codebook
Precoding matrix to be Precoding matrix used in to be used in PMI
transmission by retransmission index PDCCH without PDCCH 0
.function. ##EQU00064## -- 1 .function. ##EQU00065## -- 2
.function. ##EQU00066## -- 3 .function. ##EQU00067## -- 4
.function. ##EQU00068## -- 5 .function. ##EQU00069## -- 6
.function. ##EQU00070## .function. ##EQU00071## 7 .function.
##EQU00072## .function. ##EQU00073## 8 .function. ##EQU00074##
.function. ##EQU00075## 9 .function. ##EQU00076## .function.
##EQU00077## 10 .function. ##EQU00078## .function. ##EQU00079## 11
.function. ##EQU00080## .function. ##EQU00081##
Table 4 shows specific examples of determining retransmission
codebooks according to the fourth embodiment of the present
invention. Examples in Table 4 are given to determine which
precoding a PMI value specified in a PDCCH indicates when a UE has
two transmission antennas. The wording `precoding matrix to be used
in transmission by a PDCCH` includes (refers to) a precoding matrix
used in initial transmission of a PUSCH, or in retransmission of a
PUSCH by a PDCCH. Because a PMI is specified in a PDCCH, in
transmission of a PUSCH by a PDCCH, a UE may determine from the PMI
value which precoding matrix it should use, no matter whether the
PUSCH is in initial transmission or retransmission. The wording
`precoding matrix to be used in retransmission without transmission
of a PDCCH` refers to a precoding matrix used when retransmission
for one TB is requested only with a PHICH indicating (ACK, NACK) or
(NACK, ACK), without transmission of a PDCCH specifying a PMI. In
the examples of Table 4, a retransmission codebook is defined as
one precoding matrix in a precoding matrix to be used in
retransmission without PDCCH.
For example, assume that initial transmission of a PUSCH using a
PMI #3 was indicated by a PDCCH. A UE transmits a PUSCH by using a
matrix A as a precoding matrix. Because this PUSCH transmission was
rank-1 transmission, the UE transmits only one TB. Therefore, if an
ACK occurs, retransmission is not required, and if a NACK occurs,
the UE can use the conventional matrix A as a retransmission
precoding matrix. The reason why no precoding matrix to be used in
retransmission without PDCCH is defined for PMI #0.about.#5 in
Table 4 is because no PHICH response such as (ACK, NACK) and (NACK,
ACK) is defined since initial transmission was one-TB transmission,
for these PMIs.
As another example, assume in Table 4 that a PDCCH has indicated
initial transmission of a PUSCH using a PMI #8. A UE transmits a
PUSCH using a matrix B as a precoding matrix. Since this PUSCH
transmission was rank-2 transmission, the UE transmits two TBs. If
(ACK, ACK) occur, retransmission is not required, and if (NACK,
NACK) occur, the UE can use the matrix B as a retransmission
precoding matrix. Upon receiving a PHICH such as (ACK, NACK) and
(NACK, ACK), the UE needs to retransmit one TB, and uses a matrix C
as a rank-1 precoding matrix.
FIG. 18 illustrates a precoding determining method performed in a
Node B during retransmission, according to the fourth embodiment of
the present invention.
The procedure of FIG. 18 is assumed to be performed in step 921 of
FIG. 9. It is to be noted that a method of determining (assuming) a
precoding scheme used by a UE during retransmission by a Node B in
steps 1805 to 1821 in FIG. 18 is identical to the method of
determining (defining) a precoding scheme by a UE in steps 1705 to
1721 in FIG. 17. Since steps 1801 to 1817 in FIG. 18 are identical
in operation to steps 1201 to 1217 in FIG. 12, a detailed
description thereof will be omitted.
Referring to FIG. 18, in Case 1-1, where it is determined in step
1811 that decoding results on two TBs are (ACK, NACK) and an
initial rank was 2, a Node B determines (assumes) in step 1813 that
a UE uses a precoding matrix in an A-type rank-1 retransmission
codebook. As described above, the A-type rank-1 retransmission
codebook is indicated in an initial grant. In Case 1-2, where it is
determined in step 1811 that decoding results on two TBs are (ACK,
NACK) and an initial rank exceeds 2, the Node B determines
(assumes) in step 1815 that a UE uses a precoding matrix in an
A-type rank-2 retransmission codebook. Likewise, as described
above, the A-type rank-2 retransmission codebook is indicated in an
initial grant.
In Case 2-1, where it is determined in step 1817 that decoding
results on two TBs are (NACK, ACK) and an initial rank is less than
4, the Node B determines (assumes) in step 1819 that a UE uses a
precoding matrix in a B-type rank-1 retransmission codebook. As
described above, the B-type rank-1 retransmission codebook is
indicated in an initial grant. In Case 2-2, where it is determined
in step 1817 that decoding results on two TBs are (NACK, ACK) and
an initial rank is 4, the Node B determines (assumes) in step 1821
that a UE uses a precoding matrix in a B-type rank-2 retransmission
codebook. Likewise, as described above, the B-type rank-2
retransmission codebook is indicated in an initial grant.
The A-type retransmission codebook is used for retransmission of a
second TB, while the B-type retransmission codebook is used for
retransmission of a first TB. The method of selecting a precoding
matrix in the A-type retransmission codebook and the B-type
retransmission codebook by a Node B follows any one of the rule (i)
of selecting a precoding matrix according to the RV value and the
rule (ii) of selecting a precoding matrix according to the system
frame number or the subframe number.
The basic concept of a fifth embodiment of the present invention is
that a precoding matrix to be used in retransmission is indicated
by a PHICH. In the conventional method, a PHICH indicates ACK/NACK
information. Before introduction of UL MIMO, a PHICH is a physical
layer channel indicating 1-bit ACK/NACK information. The best way
to code 1-bit information is repetition coding. However, due to the
introduction of UL MIMO, 2-bit information should be transmitted by
a PHICH because ACK/NACK information for two TBs should be
provided. Because of the increase in the PHICH information from 1
bit to 2 bits, in the fifth embodiment of the present invention,
the PHICH is used not only to indicate ACK/NACK information, but
also to indicate a codebook to be used in retransmission by (ACK,
NACK)/(NACK, ACK) information, increasing the amount of
information. For example, if a PHICH for supporting UL MIMO is
designed to include 3-bit information, retransmission precoding
matrix indication using a PHICH may be supported as shown in Table
5 below.
TABLE-US-00005 TABLE 5 Examples of Precoding Matrix Indication
Using 3-bit PHICH State Information State Information 000 (ACK,
ACK) 100 .times..times..times..times..function. ##EQU00082## 001
.times..times..times..times..function. ##EQU00083## 101
.times..times..times..times..function. ##EQU00084## 010
.times..times..times..times..function. ##EQU00085## 110
.times..times..times..times..function. ##EQU00086## 011
.times..times..times..times..function. ##EQU00087## 111 (N ACK,
NACK)
Examples in Table 5 are given to specifically determine which
precoding matrix is to be used in retransmission, by a PHICH.
Specifically, Table 5 shows examples in which a retransmission
codebook indicated by a PHICH includes one precoding matrix. The
number of precoding matrixes defined in a retransmission codebook
indicated by a PHICH may be singular as in the examples of Table 5,
or may be plural. If only one precoding matrix is defined in a
retransmission codebook, only the precoding matrix can be used in
retransmission. Otherwise, if plural precoding matrixes are defined
in a retransmission codebook, a precoding matrix is selected and
used according to a specific rule in retransmission. As to a rule
of selecting a precoding matrix in retransmission, the following
rules mentioned in the second embodiment may be used.
i. Selecting a precoding matrix according to an RV value; and
ii. Selecting a precoding matrix according to a system frame number
or a subframe number
The advantages of the fifth embodiment of the present invention are
that a Node B may directly indicate a precoding matrix using a
PHICH without transmitting a PDCCH. The Node B may best know the
optional precoding matrix during retransmission, but indicating the
optimal precoding matrix during retransmission using a PDCCH
increases the required amount of resources.
FIG. 19 illustrates a precoding determining method performed in a
UE during retransmission, according to the fifth embodiment of the
present invention.
Referring to FIG. 19, a UE receives and decodes a PHICH in step
1901. Herein, a PHICH is designed as a physical layer channel
including n-bit information for indicating a precoding scheme in
retransmission, and may indicate a total of K states, where log
2(K).ltoreq.n. In step 1903, the PHICH including the n-bit
information may indicate the following multiple states according to
the PHICH states defined in this embodiment of the present
invention.
If the PHICH indicates State 1 in step 1903, the UE stops PUSCH
retransmission in step 1905, determining that ACK/NACK information
for two TBs is (ACK, ACK). In State 1, precoding information is not
required. If the PHICH indicates Case 2 in step 1903, the UE uses a
codebook A to determine a retransmission precoding matrix for two
TBs in step 1907, determining that ACK/NACK information for two TBs
is (ACK, NACK). If the PHICH indicates Case 3 in step 1903, the UE
uses a codebook B to determine a retransmission precoding matrix
for two TBs in step 1909, determining that ACK/NACK information for
two TBs is (ACK, NACK). If the PHICH indicates Case k in step 1903,
the UE uses a codebook D to determine a retransmission precoding
matrix for two TBs in step 1911, determining that ACK/NACK
information for two TBs is (NACK, ACK). Finally, if the PHICH
indicates Case K in step 1903, the UE uses a codebook Z to
determine a retransmission precoding matrix for both of a first TB
and a second TB in step 1913, determining that ACK/NACK information
for two TBs is (NACK, NACK).
In summary, using a function of an RV or a time resource number
(e.g., a system frame number and a subframe number), a precoding
matrix is selected in a retransmission codebook that is determined
according to the state information provided by a Node B using a
PHICH, and which rank transmission is performed and which TB is
transmitted in retransmission. Therefore, using this method, a
precoding matrix used during retransmission may be indicated by a
PHICH.
FIG. 20 illustrates a precoding determining method performed in a
Node B during retransmission, according to the fifth embodiment of
the present invention.
Referring to FIG. 20, a Node B receives and decodes a PUSCH in step
2001. In step 2003, the Node B determines the decoding states
indicating associated ACK/NACK information for two TBs, which were
described in steps 1905 to 1913 in FIG. 19. The decoding states are
mapped to specific codebooks that a UE uses during retransmission
of a PUSCH. In step 2005, the Node B generates a PHICH including
information indicating the decoding state and transmits the PHICH
to the UE. The information indicating the decoding state may
indicate a total of K states with n-bit information in a PHICH,
where log 2(K).ltoreq.n. A PHICH including the n-bit information
may indicate the multiple states described in FIG. 19, according to
the PHICH state defined in this embodiment of the present
invention. Therefore, a Node B may inform a UE of a precoding
matrix to be used in retransmission, using a PHICH including
information indicating the decoding state, without transmitting a
PDCCH.
A sixth embodiment of the present invention provides a method of
determining a precoding matrix to be used in retransmission when a
PHICH responding to UL MIMO transmission indicates ACK/NACK
information for two TBs with a single ACK/NACK rather, than
indicating the ACK/NACK information independently. In this
embodiment of the present invention, a Node B transmits an ACK only
when it has successfully decoded both of two TBs, and transmits a
NACK when it has failed in decoding any one of the TBs.
Specifically, even though two TBs have been transmitted by a PUSCH,
a PHICH indicates only one ACK/NACK. If a UE has received a NACK
through such a PHICH even though a Node B has successfully decoded
any one TB, the UE should retransmit both of the two TBs because it
may not determine which TB was failed to be successfully decoded.
In this embodiment of the present invention, if a NACK is received
from a PHICH, the UE cannot but consider the NACK as (NACK,
NACK).
According to conventional teachings, in the (NACK, NACK) state, the
previous precoding matrix should be used as a retransmission
precoding matrix because retransmission should be performed using
the transmission property used in previous transmission.
In this embodiment of the present invention, however, a UE performs
precoding by selecting a retransmission precoding matrix in a
predetermined default codebook, instead of using the previous
precoding matrix as a retransmission precoding matrix. The default
codebook is predefined for every rank individually. If a NACK
received through a PHICH indicates a request for UL MIMO
retransmission, a UE precodes a PUSCH by selecting one of one or
more precoding matrixes in the default codebook.
FIG. 21 illustrates a precoding determining method performed in a
UE during retransmission, according to the sixth embodiment of the
present invention.
Since the procedure performed in FIG. 21 is identical to the
conventional UE's operation described in FIG. 4 except for an
operation in step 2117 when a NACK is received from a PHICH, a
detailed description of steps 2101 to 2115 will be omitted.
If a UE has received a NACK from a PHICH without receiving a PDCCH,
the UE selects a precoding matrix in a default codebook and uses it
for PUSCH retransmission in step 2117. One or more precoding
matrixes may be defined in a default codebook. If only one
precoding matrix is defined in a default codebook, only the
precoding matrix can be used in retransmission. Otherwise, if
multiple precoding matrixes are defined in a default codebook, a
precoding matrix is selected and used according to a specific rule
in retransmission. As to a rule of selecting a precoding matrix in
retransmission, the following rules mentioned in the second
embodiment of the present invention may be used.
i. Selecting a precoding matrix according to an RV value; and
ii. Selecting a precoding matrix according to a system frame number
or a subframe number
FIG. 22 illustrates a precoding determining method performed in a
Node B during retransmission, according to the sixth embodiment of
the present invention. The procedure of FIG. 22 is assumed to be
performed in step 921 of FIG. 9.
Referring to FIG. 22, a Node B receives and decodes a PUSCH
transmitted by a UE in step 2201, and determines in step 2203
whether it has succeeded in decoding the PUSCH and also determines
whether the decoding result or ACK/NACK information transmitted to
the UE through a PHICH is an ACK or a NACK. The determination
results may be classified into two cases. In Case 1, where an ACK
is transmitted as the decoding is successful for both of two TBs,
the Node B does not require precoding information in step 2205,
assuming that the UE will stop retransmission of the PUSCH. In Case
2, where a NACK occurs for at least one of the two TBs, the Node B
determines (assumes) in step 2207 that the UE selects a precoding
matrix in a predetermined default codebook and uses it for PUSCH
retransmission. Because the Node B and the UE select a precoding
matrix in the default codebook according to the same rule, the Node
B is not required to transmit a PDCCH indicating a precoding matrix
that the UE will use during retransmission. Although not
illustrated in FIG. 22, the Node B transmits one NACK through a
PHICH when a NACK occurs for at least one of the two TBs.
Like in the sixth embodiment of the present invention, in a seventh
embodiment of the present invention, a PHICH provides only one
ACK/NACK. While a default codebook is defined in the sixth
embodiment of the present invention regardless of the precoding
matrix used by a UE in initial transmission, a retransmission
codebook is defined in the seventh embodiment of the present
invention according to which precoding matrix was used in initial
transmission.
One or more precoding matrixes may be defined in the retransmission
codebook of the seventh embodiment of the present invention. If
only one precoding matrix is defined in the retransmission
codebook, only the precoding matrix can be used in retransmission.
Otherwise, if multiple precoding matrixes are defined in the
retransmission codebook, a precoding matrix is selected and used
according to a predetermined rule in retransmission. As to a rule
of selecting a precoding matrix in retransmission, the following
rules mentioned in the second embodiment of the present invention
may be used.
i. Selecting a precoding matrix according to an RV value; and
ii. Selecting a precoding matrix according to a system frame number
or a subframe number
Even the case where the retransmission codebook is fully coincident
with each rank-specific codebook may be an example of the seventh
embodiment of the present invention. For example, if a PMI #p of a
rank-r is assumed to be designated as a precoding matrix by a PDCCH
using an initial grant, the PMI #p of a rank-r is used as a
precoding matrix in retransmission. It may be assumed that q=f(p),
q=f(p, RV), q=f(p, k), q=f(p, n) or q=f(p, n.sub.SFN), where n
represents a subframe number, n.sub.SFN represents a system frame
number, and k represents the value defined in Equation (4)
above.
Equation (5) below shows a simple example of q=f(p).
q=f(p)=mod(p+1,P.sub.r) (5) P.sub.r represents a size of a rank-r
codebook. P.sub.1=6 and P.sub.2=1 in Table 2 with two transmission
antennas, and P.sub.1=24, P.sub.2=16, P.sub.3=12, and P.sub.4=1 in
Table 3 with four transmission antennas.
FIG. 23 illustrates a precoding determining method performed in a
UE during retransmission, according to an eighth embodiment of the
present invention.
Referring to FIG. 23, a UE receives and decodes a PHICH in step
2301, and determines in step 2303 whether ACK/NACK information
carried by the PHICH is an ACK. The determination results may be
classified into three cases. In Case 1, where an ACK is received
for both of two TBs, the UE stops retransmission of a PUSCH and
does not require precoding information in step 2305. In Case 2,
where an ACK is received for one TB but a NACK is received for the
other TB, the UE specifically determines for which TB it has
received an ACK in step 2309. In Case 3, where a NACK is received
for both of two TBs, the UE determines to reuse the precoding
indicated in an initial grant intact in retransmission in step
2307.
In step 2309, the UE determines for which TB it has received an
ACK, when it has received an ACK for one TB but a NACK for the
other TB. In Case 1, where (ACK, NACK) are identified for TBs, it
is assumed that an ACK is identified for TB1 and a NACK is
identified for TB2. In this case, the UE performs step 2311 and its
succeeding steps. In Case 2, where (NACK, ACK) are identified for
TBs, it is assumed that a NACK is identified for TB1 and an ACK is
identified for TB2. In this case, the UE performs step 2317 and its
succeeding steps.
Step 2311 corresponds to a process of determining which rank was
used in an initial transmission of a PUSCH in Case 1. An initial
rank is always greater than or equal to 2, because it is assumed
herein that in initial transmission, two TBs are transmitted by UL
MIMO. In Case 1-1, where it is determined in step 2311 that an
initial rank is 2, the UE performs step 2313. In Case 1-2, where an
initial rank exceeds 2, the UE performs step 2315. In step 2313,
the UE implicitly determines a first column vector in a precoding
matrix used in initial transmission as a retransmission precoding
matrix without reception of a PDCCH. In step 2315, the UE
determines the first two column vectors in a precoding matrix used
in initial transmission as a retransmission precoding matrix.
In Case 2, where it is determined in step 2309 that ACK/NACK
information carried by the PHICH is identified as (NACK, ACK), the
UE determines which rank was used in initial transmission of a
PUSCH in step 2317. An initial rank is always greater than or equal
to 2, because it is assumed herein that in initial transmission,
two TBs are transmitted by UL MIMO. In Case 2-1, where it is
determined in step 2317 that an initial rank is less than 4, the UE
performs step 2313. In Case 2-2, where an initial rank is 4, the UE
performs step 2315. If a precoding matrix is determined in step
2313 or 2315, the PUSCH is retransmitted in step 2321. As to a
method of determining power used for the PUSCH transmission, there
are two possible methods. A first method is to increase
transmission power by 3 dB because of the decrease in transmission
rank, and a second method is to maintain transmission power to
reduce interference to other users.
In the eighth embodiment of the present invention, the method of
determining a precoding matrix in the (ACK, NACK) or (NACK, ACK)
state is based on the number of layers over which the TB required
to be retransmitted was transmitted in initial transmission. In
Cases 1-1 and 2-1, where the retransmission TB was transmitted over
one layer in initial transmission, the UE implicitly determines a
first column vector in a precoding matrix used in initial
transmission as a retransmission precoding matrix without reception
of a PDCCH. On the other hand, in Cases 1-2 and 2-2, where the
retransmission TB was transmitted over two layers in initial
transmission, the UE implicitly determines the first two column
vectors in a precoding matrix used in initial transmission as a
retransmission precoding matrix without reception of a PDCCH.
FIG. 24 illustrates a precoding determining method performed in a
Node B during retransmission, according to the eighth embodiment of
the present invention. The procedure of FIG. 24 is assumed to be
performed in step 921 of FIG. 9. It is to be noted that a method of
determining (assuming) a precoding scheme used by a UE during
retransmission by a Node B in steps 2405 to 2415 in FIG. 24 is
identical to the method of determining (defining) a precoding
scheme by a UE in steps 2305 to 2315 in FIG. 23.
Referring to FIG. 24, a Node B receives and decodes a PUSCH
transmitted by a UE in step 2401, and determines in step 2403
whether the decoding of the PUSCH is successful and also determines
whether the decoding result or ACK/NACK information transmitted to
the UE through a PHICH is an ACK or a NACK. The determination
results may be classified into three cases. In Case 1, where it is
determined, as in step 909 of FIG. 9, that an ACK is transmitted
for both of two TBs because of the success in decoding for both of
two TBs, the Node B assumes in step 2405 that the UE will stop
retransmission of the PUSCH, and does not require precoding
information. In Case 2, where it is determined, as in steps 911 and
913 of FIG. 9, that an ACK is transmitted for one TB but a NACK is
transmitted for the other TB, the Node B specifically determines
for which TB it will transmit an ACK in step 2409, and performs its
succeeding steps. In Case 3, where it is determined, as in step
915, of FIG. 9 that a NACK is transmitted for both of two TBs
because of the failure in decoding for both of two TBs, the Node B
assumes in step 2407 that the UE reuses the precoding indicated in
an initial grant, in retransmission.
In step 2409, the Node B determines for which TB it has transmitted
an ACK in Case 2 where an ACK was transmitted for one TB but a NACK
was transmitted for the other TB. In Case 1, where (ACK, NACK) are
identified for TBs, it is assumed that an ACK is identified for TB1
and a NACK is identified for TB2. In this case, the Node B performs
step 2411 and its succeeding steps. In Case 2, where (NACK, ACK)
are identified for TBs, it is assumed that a NACK is identified for
TB1 and an ACK is identified for TB2. In this case, the Node B
performs step 2417 and its succeeding steps.
Step 2411 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 1 where ACK/NACK
information is identified as (ACK, NACK) in step 2409. An initial
rank is always greater than or equal to 2, because it is assumed
herein that in initial transmission, two TBs are transmitted by UL
MIMO. In Case 1-1, where it is determined in step 2411 that an
initial rank is 2, the Node B performs step 2413, and in Case 1-2,
where an initial rank exceeds 2, the Node B performs step 2415. In
step 2413, the Node B implicitly determines (assumes) a first
column vector in a precoding matrix used in initial transmission as
a retransmission precoding matrix used by the UE. In step 2415, the
Node B implicitly determines (assumes) the first two column vectors
in a precoding matrix used in initial transmission as a
retransmission precoding matrix used by the UE.
In Case 2 where it is determined in step 2409 that ACK/NACK
information is identified as (NACK, ACK), the Node B determines
which rank was used in initial transmission of a PUSCH in step
2417. An initial rank is always greater than or equal to 2, because
it is assumed herein that in initial transmission, two TBs are
transmitted by UL MIMO. In Case 2-1, where it is determined in step
2417 that an initial rank is less than 4, the Node B performs step
2413, and in Case 2-2, where an initial rank is 4, the Node B
performs step 2415.
In the eighth embodiment of the present invention, each of the Node
B and the UE implicitly determines the first one or two column
vectors in a precoding matrix used in initial transmission as a
retransmission precoding matrix. In the alternative, however, the
UE may be modified to implicitly determine the last one column
vector in a precoding matrix used in initial transmission as a
retransmission precoding matrix without reception of a PDCCH in
Cases 1-1 and 2-1 where a transmission TB was transmitted over one
layer in initial transmission, and to implicitly determine the last
two column vectors in a precoding matrix used in initial
transmission as a retransmission precoding matrix without reception
of a PDCCH in Cases 1-2 and 2-2 where a transmission TB was
transmitted over two layers in initial transmission.
FIG. 25 illustrates a precoding determining method performed in a
UE during retransmission, according to a ninth embodiment of the
present invention.
Referring to FIG. 25, a UE receives and decodes a PHICH in step
2501, and determines in step 2503 whether ACK/NACK information
carried by the PHICH is an ACK. The determination results may be
classified into three cases. In Case 1, where an ACK is received
for both of two TBs, the UE stops retransmission of a PUSCH and
does not require precoding information in step 2505. In Case 2,
where an ACK is received for one TB but a NACK is received for the
other TB, the UE specifically determines for which TB it has
received an ACK in step 2509, and performs its succeeding steps. In
Case 3, where a NACK is received for both of two TBs, the UE
determines to reuse the precoding indicated in an initial grant, in
retransmission in step 2507.
In step 2509, the UE determines for which TB it has received an ACK
in Case 2 where an ACK was received for one TB but a NACK was
received for the other TB. In Case 1, where (ACK, NACK) are
identified for TBs, it is assumed that an ACK is identified for TB1
and a NACK is identified for TB2. In this case, the UE performs
step 2511 and its succeeding steps. In Case 2, where (NACK, ACK)
are identified for TBs, it is assumed that a NACK is identified for
TB1 and an ACK is identified for TB2. In this case, the UE performs
step 2521 and its succeeding steps.
Step 2511 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 1 where it is
determined in step 2509 that ACK/NACK information carried by the
PHICH is identified as (ACK, NACK). An initial rank is always
greater than or equal to 2, because it is assumed herein that in
initial transmission, two TBs are transmitted by UL MIMO. In Case
1-1, where it is determined in step 2511 that an initial rank is 2,
the UE performs step 2513, and in Case 1-2, where an initial rank
exceeds 2, the UE performs step 2523.
In step 2513, the UE determines if an MCS level of TB1 is higher
than an MCS level of TB2. If the MCS level of TB1 is higher,
meaning that a channel state of the layer used by TB1 is better,
then the UE performs step 2515. Otherwise, the UE performs step
2517. In the alternative, it may be determined in step 2513 whether
an MCS level of TB1 is not lower than an MCS level of TB2.
In step 2515, the UE implicitly determines a first column vector in
a precoding matrix used in initial transmission as a retransmission
precoding matrix without reception of a PDCCH. The first column
vector was used by TB1. This is to use the layer of TB1 even though
TB2 is retransmitted because the channel state of the layer used by
TB1 is better.
In step 2517, the UE implicitly determines the last column vector
in a precoding matrix used in initial transmission as a
retransmission precoding matrix without reception of a PDCCH. The
last column vector was originally used by TB2. This is to reuse the
last column vector as a retransmission precoding matrix because the
channel state of the layer used by TB2 is better.
Similarly, in step 2523, the UE determines if an MCS level of TB1
is higher than an MCS level of TB2. If the MCS level of TB1 is
higher, meaning that a channel state of the layer used by TB1 is
better, then the UE performs step 2525. Otherwise, the UE performs
step 2527. In the alternative, it may be determined in step 2523
whether an MCS level of TB1 is not lower than an MCS level of
TB2.
In step 2525, the UE implicitly determines the first two column
vectors in a precoding matrix used in initial transmission as a
retransmission precoding matrix without reception of a PDCCH. The
first one or two column vectors were used by TB1. This is to use
the layer of TB1 even though TB2 is retransmitted because the
channel state of the layer used by TB1 is better.
In step 2527, the UE implicitly determines the last two column
vectors in a precoding matrix used in initial transmission as a
retransmission precoding matrix without reception of a PDCCH. The
last one or two column vectors were originally used by TB2. This is
to reuse the last two column vectors as a retransmission precoding
matrix because the channel state of the layer used by TB2 is
better.
Step 2521 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 2 where it is
determined in step 2509 that ACK/NACK information carried by the
PHICH is identified as (NACK, ACK). An initial rank is always
greater than or equal to 2, because it is assumed herein that in
initial transmission, two TBs are transmitted by UL MIMO. In Case
2-1, where it is determined in step 2521 that an initial rank is
less than 4, the UE performs step 2513, and in Case 2-2, where an
initial rank is 4, the UE performs step 2523.
After a precoding matrix is determined in step 2515 or 2517, the
PUSCH is retransmitted in step 2519. Likewise, after a precoding
matrix is determined in step 2525 or 2527, the PUSCH is
retransmitted in step 2519. As to a method of determining power
used for the PUSCH transmission, there are two possible methods. A
first method is to increase transmission power by 3 dB because of
the decrease in transmission rank, and a second method is to
maintain transmission power to reduce interference to other
users.
FIG. 26 illustrates a precoding determining method performed in a
Node B during retransmission, according to the ninth embodiment of
the present invention. The procedure of FIG. 26 is assumed to be
performed in step 921 of FIG. 9. It is to be noted that a method of
determining (assuming) a precoding scheme used by a UE during
retransmission by a Node B in steps 2605 to 2625 in FIG. 26 is
identical to the method of determining (defining) a precoding
scheme by a UE in steps 2505 to 2527 in FIG. 25.
Referring to FIG. 26, a Node B receives and decodes a PUSCH
transmitted by a UE in step 2601, and determines in step 2603
whether the decoding of the PUSCH is successful and also determines
whether the decoding result or ACK/NACK information transmitted to
the UE through a PHICH is an ACK or a NACK. The determination
results may be classified into three cases. In Case 1, where it is
determined, as in step 909 of FIG. 9, that an ACK is transmitted
for both of two TBs because of the success in decoding for both of
two TBs, the Node B assumes in step 2605 that the UE will stop
retransmission of the PUSCH, and does not require precoding
information. In Case 2, where it is determined, as in steps 911 and
913 of FIG. 9, that an ACK is transmitted for one TB but a NACK is
transmitted for the other TB, the Node B specifically determines
for which TB it will transmit an ACK in step 2609, and performs its
succeeding steps. In Case 3, where it is determined, as in step 915
of FIG. 9, that a NACK is transmitted for both of two TBs because
of the failure in decoding for both of two TBs, the Node B assumes
in step 2607 that the UE reuses the precoding indicated in an
initial grant, in retransmission.
In step 2609, the Node B determines for which TB it has transmitted
an ACK in Case 2 where an ACK was transmitted for one TB but a NACK
was transmitted for the other TB. In Case 1, where (ACK, NACK) are
identified for TBs, it is assumed that an ACK is identified for TB1
and a NACK is identified for TB2. In this case, the Node B performs
step 2611 and its succeeding steps. In Case 2, where (NACK, ACK)
are identified for TBs, it is assumed that a NACK is identified for
TB1 and an ACK is identified for TB2. In this case, the Node B
performs step 2619 and its succeeding steps.
Step 2611 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 1 where ACK/NACK
information is identified as (ACK, NACK) in step 2609. An initial
rank is always greater than or equal to 2, because it is assumed
herein that in initial transmission, two TBs are transmitted by UL
MIMO. In Case 1-1, where it is determined in step 2611 that an
initial rank is 2, the Node B performs step 2613, and in Case 1-2,
where an initial rank exceeds 2, the Node B performs step 2621.
In step 2613, the Node B determines if an MCS level of TB1 is
higher than an MCS level of TB2. If the MCS level of TB1 is higher,
the Node B performs step 2615. Otherwise, the Node B performs step
2617. In the alternative, it may be determined in step 2613 whether
an MCS level of TB1 is not lower than an MCS level of TB2.
In step 2615, the Node B implicitly determines (assumes) a first
column vector in a precoding matrix used in initial transmission as
a retransmission precoding matrix used by the UE. On the other
hand, in step 2617, the Node B implicitly determines (assumes) the
last column vector in a precoding matrix used in initial
transmission as a retransmission precoding matrix used by the
UE.
Similarly, in step 2621, the Node B determines if an MCS level of
TB1 is higher than an MCS level of TB2. If the MCS level of TB1 is
higher, the Node B performs step 2623. Otherwise, the Node B
performs step 2625. In the alternative, it may be determined in
step 2621 whether an MCS level of TB1 is not lower than an MCS
level of TB2.
In step 2623, the Node B implicitly determines (assumes) the first
two column vectors in a precoding matrix used in initial
transmission as a retransmission precoding matrix used by the UE.
On the other hand, in step 2625, the Node B implicitly determines
(assumes) the last two column vectors in a precoding matrix used in
initial transmission as a retransmission precoding matrix used by
the UE.
Step 2619 corresponds to a process of determining which rank was
used in initial transmission of a PUSCH in Case 2 where it is
determined in step 2609 that ACK/NACK information carried by the
PHICH is identified as (NACK, ACK). An initial rank is always
greater than or equal to 2, because it is assumed herein that in
initial transmission, two TBs are transmitted by UL MIMO. In Case
2-1, where it is determined in step 2619 that an initial rank is
less than 4, the UE performs step 2613, and in Case 2-2, where an
initial rank is 4, the UE performs step 2621.
A tenth embodiment of the present invention is a combination of the
third and fourth embodiments of the present invention. In the third
embodiment of the present invention, with the introduction of the
mother-child pair concept, a precoding matrix to be used in
retransmission is determined by the precoding matrix used in
initial transmission. In the fourth embodiment of the present
invention, an initial grant is designated not only the precoding
matrix to be used in initial transmission, but also the precoding
matrix to be used in retransmission without transmission of a
PDCCH. In the tenth embodiment of the present invention, a child
codebook is defined by the precoding matrix used in initial
transmission as in the third embodiment of the present invention,
and a precoding matrix to be used in retransmission without
transmission of a PDCCH is designated in the child codebook as in
the fourth embodiment of the present invention. In this case,
multiple precoding matrixes are provided in the child codebook.
In all of the above-described embodiments of the present invention,
it is assumed that retransmission is indicated by a PHICH. The
following example is about a method of applying the above
embodiments when indicated by a PDCCH for fallback. PDCCH-resource
consumption may be significantly reduced by using a DCI format
designed considering single-antenna transmission by a Node B
without using UL MIMO. This is because a DCI format (e.g., DCI
format 0) designed considering single-antenna transmission is less
than a UL MIMO DCI format in terms of the amount of information due
to the non-necessity of expressing multiple TBs and PMIs. The term
`fallback` refers to sending a DCI to a UE whose channel state
becomes poor all of a sudden. As for a DCI whose information is
reduced significantly, even a UE with a poor channel state may
likely receive the DCI, because the DCI uses less resource.
Even though retransmission is indicated by a PDCCH, if it is based
on a DCI format 0, PMI information may not be delivered. In this
case, a method of determining a precoding matrix in retransmission
may be defined by utilizing the above embodiments of the present
invention. To be specific, when retransmission is requested by the
DCI format 0, precoding is performed using a predetermined
precoding matrix or a precoding matrix that is selected in a
retransmission codebook or a candidate precoding matrix group
according to a specific rule. A method of defining a retransmission
precoding matrix or a retransmission codebook follows the
embodiments of the present invention.
In the embodiments of the present invention, one of a method of
using a codebook in which rank-1 precoding matrixes are defined and
a method of using a codebook in which rank-2 precoding matrixes are
defined, has been considered according to which TB is
retransmitted. However, the change in rank depending on the channel
state may be insignificant. In this case, given that a rank-2
precoding matrix may be applied even in retransmission if rank-r
(where r.gtoreq.2) could be supported in initial transmission, it
may be preferable to apply a rank-2 precoding matrix rather than a
rank-1 precoding matrix in retransmission of the TB that has
occupied one layer in initial transmission. Therefore, the above
embodiments may be modified to apply a precoding matrix selected in
a rank-2 codebook even to a retransmission TB regardless of which
TB is retransmitted.
For example, in the second embodiment of the present invention, a
default rank-1 codebook is replaceable with a default rank-2
codebook. In the third embodiment of the present invention, a
rank-1 child codebook is replaceable with a rank-2 child codebook.
In the fourth embodiment of the present invention, a rank-1
codebook may be replaced with a rank-2 codebook.
While the invention has been shown and described with reference to
certain embodiments thereof, it will be understood by those skilled
in the art that various changes in form and detail may be made
therein without departing from the spirit and scope of the
invention as defined by the appended claims and their
equivalents.
* * * * *